INDEX

Explanations

positive statements or actions

oai_token-act-pair · gpt-3.5-turbo

conjunctions and phrases that emphasize additional points or comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

"},"

-0.78

 ÂŃ

-0.62

remlin

-0.61

ventures

-0.61

=~=~

-0.60

icipated

-0.60

kefeller

-0.60

natureconservancy

-0.58

 âĢİ

-0.57

prus

-0.55

POSITIVE LOGITS

adays

1.19

1.10

,.

1.09

,...

0.94

,,

0.89

 beware

0.88

we

0.85

if

0.84

,-

0.83

 thats

0.81

Activations Density 0.274%

positive statements or actions

conjunctions and phrases that emphasize additional points or comparisons

No Comments

No Known Activations