INDEX

Explanations

phrases related to political positions or actions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating future actions or expectations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-0.94

MpServer

-0.88

abiding

-0.79

breeding

-0.77

processing

-0.76

 Occupations

-0.76

anooga

-0.75

imore

-0.74

pattern

-0.72

 Printed

-0.69

POSITIVE LOGITS

 resign

1.60

 appoint

1.55

 veto

1.38

 nominate

1.35

 oversee

1.28

 announce

1.27

 propose

1.27

 personally

1.23

 rescind

1.19

 endorse

1.18

Activations Density 0.270%

phrases related to political positions or actions

phrases indicating future actions or expectations

No Comments

No Known Activations