INDEX

Explanations

words related to justification or reasoning for actions, decisions, or opinions

oai_token-act-pair · gpt-3.5-turbo

terms related to justification and rationalization

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ummer

-0.82

ammy

-0.76

grass

-0.72

berry

-0.71

chron

-0.70

ept

-0.69

jong

-0.69

 Carbuncle

-0.66

enfranch

-0.66

ker

-0.65

POSITIVE LOGITS

 justifying

1.07

 justifies

0.99

 justify

0.98

 justification

0.97

why

0.82

 excuses

0.81

¿½

0.76

WHY

0.76

 justified

0.75

ighed

0.73

Activations Density 0.032%

words related to justification or reasoning for actions, decisions, or opinions

terms related to justification and rationalization

No Comments

No Known Activations