INDEX

Explanations

date ranges

oai_token-act-pair · gpt-3.5-turbo

phrases indicating time periods or temporal references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

airy

-0.79

aeus

-0.69

achy

-0.67

cit

-0.66

etric

-0.64

ocus

-0.62

inant

-0.61

DEF

-0.61

ysis

-0.61

IGHT

-0.60

POSITIVE LOGITS

RTX

0.69

 Governors

0.64

bury

0.63

avez

0.62

 checkpoints

0.61

 Palo

0.60

 Tibet

0.60

 Shanghai

0.60

 Chef

0.59

Activations Density 0.194%

date ranges

phrases indicating time periods or temporal references

No Comments

No Known Activations