INDEX

Explanations

verbs that indicate taking a specific action or making a specific decision

oai_token-act-pair · gpt-3.5-turbo

references to decisions and their implications

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

staff

-0.65

enne

-0.61

 Fact

-0.60

ennes

-0.60

umb

-0.59

created

-0.58

hn

-0.58

rising

-0.57

verages

-0.56

opened

-0.55

POSITIVE LOGITS

ylum

0.73

 mistake

0.67

olate

0.66

 disappear

0.65

ãĥĥãĤ¯

0.65

vow

0.65

kish

0.65

 impression

0.64

 vanish

0.64

 contribution

0.63

Activations Density 0.193%

verbs that indicate taking a specific action or making a specific decision

references to decisions and their implications

No Comments

No Known Activations