INDEX

Explanations

the word "het" with varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

references to specific substances or drugs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 eternal

-0.61

 rampant

-0.60

STAR

-0.60

 kindred

-0.59

reverse

-0.58

BACK

-0.58

equ

-0.56

 viewing

-0.56

ittal

-0.56

 temp

-0.55

POSITIVE LOGITS

amines

1.11

orically

1.02

orical

1.00

ting

0.98

icals

0.98

ty

0.97

oric

0.97

ype

0.96

ted

0.95

ilde

0.93

Activations Density 0.027%

the word "het" with varying levels of activation

references to specific substances or drugs

No Comments

No Known Activations