INDEX

Explanations

words related to intense heat, fire, and speed

oai_token-act-pair · gpt-3.5-turbo

terms related to fire or heat

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

pai

-0.83

itol

-0.73

icho

-0.70

entary

-0.67

ajor

-0.67

ropy

-0.66

 coerc

-0.64

ulates

-0.63

 Territories

-0.62

anyl

-0.62

POSITIVE LOGITS

bird

0.81

ahead

0.75

onde

0.75

 blaze

0.74

birds

0.74

geoning

0.74

EED

0.74

0.73

quit

0.71

hawk

0.69

Activations Density 0.045%

words related to intense heat, fire, and speed

terms related to fire or heat

No Comments

No Known Activations