INDEX

Explanations

mentions of specific organizations or individuals who had been involved in significant events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "had."

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

hammer

-0.67

hack

-0.63

bie

-0.61

âĻ¦

-0.60

porting

-0.60

 Voters

-0.59

externalActionCode

-0.58

 Skies

-0.57

eem

-0.56

vt

-0.56

POSITIVE LOGITS

 been

1.33

 undergone

1.18

 begun

1.11

 gone

1.05

iths

1.02

 gotten

0.96

been

0.93

 taken

0.92

 become

0.89

 flown

0.87

Activations Density 0.159%

mentions of specific organizations or individuals who had been involved in significant events

occurrences of the word "had."

No Comments

No Known Activations