INDEX

Explanations

phrases related to directional movement or transition

oai_token-act-pair · gpt-3.5-turbo

instances of the word "into" and related context

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 icing

-0.64

due

-0.64

 opinion

-0.62

 sessions

-0.61

 parties

-0.61

DD

-0.61

 eval

-0.60

 chat

-0.60

 breakout

-0.60

 Warm

-0.59

POSITIVE LOGITS

into

3.70

onto

1.01

 Into

1.00

inside

0.97

hiba

0.82

lda

0.82

ever

0.82

indu

0.81

 INTO

0.80

INT

0.79

Activations Density 0.009%

phrases related to directional movement or transition

instances of the word "into" and related context

No Comments

No Known Activations