INDEX

Explanations

words related to justification or reasoning for actions

oai_token-act-pair · gpt-3.5-turbo

terms related to justification and reasoning

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

ummer

-0.84

 Carbuncle

-0.76

chron

-0.72

ammy

-0.72

Hop

-0.71

estone

-0.70

berry

-0.69

ept

-0.68

grass

-0.68

thritis

-0.67

POSITIVE LOGITS

 justifying

1.00

 justification

0.89

 justify

0.87

 justifies

0.87

why

0.79

ably

0.74

 justified

0.73

 excuses

0.72

 guiActiveUn

0.70

ighed

0.69

Activations Density 0.032%

words related to justification or reasoning for actions

terms related to justification and reasoning

No Comments

No Known Activations