INDEX

Explanations

numerical values, specifically year-related information

oai_token-act-pair · gpt-3.5-turbo

numerical data and statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gets

-0.75

uden

-0.74

tons

-0.73

faced

-0.72

mes

-0.70

journal

-0.69

matically

-0.69

lying

-0.66

mits

-0.66

lie

-0.64

POSITIVE LOGITS

0.89

IRC

0.88

uador

0.82

REM

0.80

ASC

0.79

Â¯Â¯Â¯Â¯

0.75

0.74

eenth

0.74

ength

0.73

Activations Density 0.047%

numerical values, specifically year-related information

numerical data and statistics

No Comments

No Known Activations