INDEX

Explanations

the word "pose" or its variations, specifically in the context of suggesting a threat or influence

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of posing or presentation

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

 miracles

-0.64

 divorce

-0.60

LINE

-0.59

ann

-0.59

 books

-0.59

 disinteg

-0.58

 rushes

-0.58

 miles

-0.57

 graduating

-0.56

 wins

-0.56

POSITIVE LOGITS

pose

4.18

posed

3.46

posing

3.30

poses

3.27

position

1.90

POSE

1.86

pos

1.84

posal

1.56

POS

1.55

posure

1.39

Activations Density 0.007%

the word "pose" or its variations, specifically in the context of suggesting a threat or influence

words related to the concept of posing or presentation

No Comments

No Known Activations