INDEX

Explanations

occurrences of the word "after" followed by something else

oai_token-act-pair · gpt-3.5-turbo

phrases indicating repetition or frequent occurrences of events

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

ueller

-0.74

:[

-0.71

ollar

-0.69

ortium

-0.67

hirt

-0.66

enhagen

-0.66

 holster

-0.62

 enshr

-0.62

odore

-0.61

ieth

-0.59

POSITIVE LOGITS

 Kinnikuman

0.82

 guiActiveUnfocused

0.78

rams

0.76

URES

0.73

CTV

0.69

 Kits

0.65

 syndrome

0.65

ements

0.63

 Thumbnails

0.63

Redd

0.62

Activations Density 1.554%

occurrences of the word "after" followed by something else

phrases indicating repetition or frequent occurrences of events

No Comments

No Known Activations