INDEX

Explanations

words related to actions or processes that involve a level of deliberation or decision-making

oai_token-act-pair · gpt-3.5-turbo

words related to various forms of "actions" or "activities."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ALL

-0.63

Present

-0.62

URRENT

-0.61

ASED

-0.60

STAT

-0.60

âĹı

-0.60

ãĥ£

-0.60

avez

-0.59

CHAR

-0.58

AGES

-0.58

POSITIVE LOGITS

ations

1.24

hower

1.17

etting

0.97

etter

0.95

leeve

0.90

eus

0.89

arily

0.86

ulations

0.85

chwitz

0.85

hip

0.84

Activations Density 0.024%

words related to actions or processes that involve a level of deliberation or decision-making

words related to various forms of "actions" or "activities."

No Comments

No Known Activations