INDEX

Explanations

words related to opposition or conflict

oai_token-act-pair · gpt-3.5-turbo

instances of the term "opposing" in various contexts

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

 Virgin

-0.66

ometime

-0.63

ps

-0.62

Scale

-0.61

pd

-0.59

 Paste

-0.59

 Blend

-0.59

flow

-0.59

 Plug

-0.59

 Scale

-0.59

POSITIVE LOGITS

 opposing

3.94

 opposition

1.73

 opponents

1.69

 oppos

1.66

 opposite

1.61

 opponent

1.59

 dissenting

1.47

Opp

1.47

 defending

1.45

 opposed

1.38

Activations Density 0.007%

words related to opposition or conflict

instances of the term "opposing" in various contexts

No Comments

No Known Activations