INDEX

Explanations

words related to actions or objects related to a specific activity

oai_token-act-pair · gpt-3.5-turbo

words related to cooking or food preparation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 microscope

-0.72

 Weir

-0.69

 Trop

-0.66

 corpor

-0.64

 backs

-0.63

Gir

-0.62

 victorious

-0.61

 strings

-0.60

 boosted

-0.60

 Socialism

-0.58

POSITIVE LOGITS

ake

4.62

akes

2.82

aking

1.89

aken

1.59

aked

1.47

akable

1.46

aker

1.38

akers

1.38

aky

1.35

ook

1.32

Activations Density 0.009%

words related to actions or objects related to a specific activity

words related to cooking or food preparation

No Comments

No Known Activations