INDEX

Explanations

phrases related to outcomes or results

oai_token-act-pair · gpt-3.5-turbo

phrases that involve the phrase "ended up."

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

ongyang

-0.78

 assurance

-0.69

mens

-0.66

mouth

-0.65

Nap

-0.63

Reviewed

-0.63

Ctrl

-0.61

 Rothschild

-0.60

ï¸ı

-0.60

bay

-0.59

POSITIVE LOGITS

adesh

0.76

 coli

0.74

skirts

0.73

 TAMADRA

0.72

ãĥĩãĤ£

0.72

 redes

0.72

acters

0.71

ymes

0.70

votes

0.70

steen

0.70

Activations Density 0.023%

phrases related to outcomes or results

phrases that involve the phrase "ended up."

No Comments

No Known Activations