INDEX

Explanations

words related to burning or fire

oai_token-act-pair · gpt-3.5-turbo

phrases related to burning or fire-related actions

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

eele

-1.12

agonists

-0.80

elvet

-0.79

incial

-0.76

CLASS

-0.74

asking

-0.71

ORTS

-0.71

ioch

-0.69

 latitude

-0.68

UA

-0.66

POSITIVE LOGITS

 fuse

0.76

 hydrogen

0.74

 sulfur

0.74

 asphalt

0.73

 contaminated

0.72

 contam

0.69

 polluted

0.69

 crem

0.69

 boil

0.67

 brim

0.67

Activations Density 0.262%

words related to burning or fire

phrases related to burning or fire-related actions

No Comments

No Known Activations