INDEX

Explanations

mentions of exceptional achievements or awards

oai_token-act-pair · gpt-3.5-turbo

words and phrases associated with awards and honors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 inconven

-0.72

 desperation

-0.71

morrow

-0.69

enes

-0.69

 frust

-0.68

 Â¯

-0.67

abal

-0.67

ropri

-0.66

 flashbacks

-0.65

 tactics

-0.64

POSITIVE LOGITS

 prestigious

1.11

ottest

1.05

 Prize

0.99

 Award

0.98

 award

0.94

 coveted

0.90

 Awards

0.89

 awards

0.88

 honors

0.87

 awarded

0.87

Activations Density 0.613%

mentions of exceptional achievements or awards

words and phrases associated with awards and honors

No Comments

No Known Activations