INDEX

Explanations

mentions of the word 'gl' with varying intensities

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "gl" or related sequences associated with global or environmental themes

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

 delinquent

-0.71

 Abdul

-0.68

 Yard

-0.67

 authorized

-0.67

 monet

-0.62

 rank

-0.62

 AUTHOR

-0.62

Nas

-0.62

 Ricardo

-0.61

 Chiefs

-0.61

POSITIVE LOGITS

gl

4.19

Gl

2.19

GL

1.94

Gl

1.66

gl

1.52

glass

1.34

GL

1.33

ogl

1.31

glas

1.25

gd

1.24

Activations Density 0.008%

mentions of the word 'gl' with varying intensities

occurrences of the term "gl" or related sequences associated with global or environmental themes

No Comments

No Known Activations