INDEX

Explanations

dates expressed in the format "month day, year"

oai_token-act-pair · gpt-3.5-turbo

dates and numerical information related to events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

everal

-0.98

senal

-0.90

 tremend

-0.90

ccording

-0.86

aditional

-0.85

 cumbers

-0.82

 streng

-0.82

ensical

-0.82

 exha

-0.82

oreAnd

-0.80

POSITIVE LOGITS

1.06

1.01

1.00

0.98

0.90

0.84

0.81

Activations Density 0.042%

dates expressed in the format "month day, year"

dates and numerical information related to events

No Comments

No Known Activations