INDEX

Explanations

dates and specific events or entities mentioned in a structured format

oai_token-act-pair · gpt-3.5-turbo

references to specific locations, items, and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Footnote

-0.61

picture

-0.58

prim

-0.56

ãĥ¼ãĥ³

-0.56

igm

-0.55

og

-0.55

ppo

-0.54

ounding

-0.54

 attRot

-0.53

otype

-0.52

POSITIVE LOGITS

at

1.77

At

1.31

at

1.27

At

1.25

AT

0.93

atography

0.91

0.68

 elsewhere

0.66

 somewhere

0.64

 whatsoever

0.62

Activations Density 0.293%

dates and specific events or entities mentioned in a structured format

references to specific locations, items, and events

No Comments

No Known Activations