INDEX

Explanations

information related to actions or events

oai_token-act-pair · gpt-3.5-turbo

key terms and phrases related to actions and decisions

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

 Mund

-0.84

MS

-0.78

-0.77

 Madison

-0.75

+++

-0.74

 Gamb

-0.73

-0.72

 Berman

-0.72

-0.71

POSITIVE LOGITS

ake

1.10

akes

1.09

 Take

1.03

take

1.02

Take

0.98

 take

0.94

amaru

0.92

 Taken

0.89

 Takes

0.88

 Draco

0.86

Activations Density 0.243%

information related to actions or events

key terms and phrases related to actions and decisions

No Comments

No Known Activations