INDEX

Explanations

verbs denoting communication such as speaking, saying, and thinking

oai_token-act-pair · gpt-3.5-turbo

verbs indicating ongoing actions or contributions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

selves

-0.78

Higher

-0.70

respective

-0.70

wayne

-0.69

ocating

-0.68

iners

-0.67

iky

-0.65

RELATED

-0.64

wik

-0.63

illion

-0.62

POSITIVE LOGITS

 himself

0.82

 herself

0.76

his

0.67

 brilliantly

0.65

 shotgun

0.63

 Bord

0.60

 unsuccessfully

0.60

 sage

0.59

 valiant

0.59

ographs

0.59

Activations Density 0.420%

verbs denoting communication such as speaking, saying, and thinking

verbs indicating ongoing actions or contributions

No Comments

No Known Activations