INDEX

Explanations

food-related instructions or cooking steps

oai_token-act-pair · gpt-3.5-turbo

punctuations and conjunctions indicating sequential steps in a recipe

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

 estranged

-0.86

 indebted

-0.79

 unexplained

-0.79

 privileged

-0.76

 emblem

-0.75

 prostitutes

-0.74

 criminally

-0.73

 affili

-0.73

 offended

-0.73

 oppressed

-0.73

POSITIVE LOGITS

 Ideally

1.09

 Depending

1.09

 Alternatively

0.97

Optional

0.95

Required

0.94

 Instructions

0.92

Assembly

0.92

 Slowly

0.92

 Once

0.90

 Usually

0.90

Activations Density 0.271%

food-related instructions or cooking steps

punctuations and conjunctions indicating sequential steps in a recipe

No Comments

No Known Activations