INDEX

Explanations

words related to stories, reports, or detailed explanations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "inside."

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

lda

-0.80

enegger

-0.78

ministic

-0.78

olk

-0.77

eday

-0.75

ãĥģ

-0.70

tle

-0.69

fortune

-0.67

vous

-0.67

yah

-0.66

POSITIVE LOGITS

vert

0.85

 parentheses

0.77

 vitro

0.71

thia

0.71

 vivo

0.69

 exerc

0.69

 bedrooms

0.68

 latex

0.68

ocent

0.67

doors

0.66

Activations Density 0.020%

words related to stories, reports, or detailed explanations

occurrences of the word "inside."

No Comments

No Known Activations