INDEX

Explanations

information related to education and achievements, particularly the act of graduating from a program or institution

oai_token-act-pair · gpt-3.5-turbo

mentions of awards and recognitions received, especially related to graduation or achievements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rir

-0.89

etsk

-0.79

iquid

-0.76

oras

-0.72

abal

-0.71

utterstock

-0.71

 frust

-0.70

arre

-0.69

hed

-0.69

ullivan

-0.69

POSITIVE LOGITS

 Medal

1.07

 honors

1.04

 prestigious

1.00

 Excellence

0.90

 honorary

0.88

 graduation

0.88

 honorable

0.81

 consecutive

0.81

 honor

0.79

 Award

0.76

Activations Density 0.591%

information related to education and achievements, particularly the act of graduating from a program or institution

mentions of awards and recognitions received, especially related to graduation or achievements

No Comments

No Known Activations