INDEX

Explanations

mentions of the word "Mint" with varying strengths of activation

oai_token-act-pair · gpt-3.5-turbo

references to the word "mint" and its various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Ake

-0.72

Magikarp

-0.66

voc

-0.63

 Airbnb

-0.60

://

-0.59

 Methodist

-0.59

---------

-0.58

HUD

-0.57

 unfolding

-0.56

 adoptive

-0.56

POSITIVE LOGITS

 Seym

1.48

ãĥ¼ãĥĨãĤ£

0.87

oufl

0.84

ulously

0.80

rats

0.79

marks

0.78

ricting

0.77

zl

0.77

itary

0.76

Ã©r

0.76

Activations Density 0.029%

mentions of the word "Mint" with varying strengths of activation

references to the word "mint" and its various contexts

No Comments

No Known Activations