INDEX

Explanations

words related to strong negative emotions or situations

oai_token-act-pair · gpt-3.5-turbo

words associated with fear or intimidation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

á

-0.81

elman

-0.80

ublic

-0.80

uties

-0.76

adr

-0.74

eva

-0.73

ritis

-0.73

bers

-0.72

elf

-0.72

glas

-0.71

POSITIVE LOGITS

ly

1.16

 unbeliev

0.92

NESS

0.83

ingly

0.79

 reptiles

0.79

ively

0.77

LY

0.76

ously

0.74

 heights

0.73

mares

0.72

Activations Density 0.016%

words related to strong negative emotions or situations

words associated with fear or intimidation

No Comments

No Known Activations