INDEX

Explanations

phrases related to regular events or routines

oai_token-act-pair · gpt-3.5-turbo

phrases related to frequency and regularity of actions or events

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

ajor

-0.77

hd

-0.69

vier

-0.66

resy

-0.65

eworks

-0.65

idon

-0.63

 Compatibility

-0.63

enfranch

-0.61

emo

-0.61

ailand

-0.61

POSITIVE LOGITS

 dips

0.80

 invariably

0.76

 imaginable

0.74

pmwiki

0.69

Decre

0.66

 churn

0.66

 updates

0.65

 nightly

0.64

 inevitably

0.64

tick

0.64

Activations Density 0.335%

phrases related to regular events or routines

phrases related to frequency and regularity of actions or events

No Comments

No Known Activations