INDEX

Explanations

words related to aggression

oai_token-act-pair · gpt-3.5-turbo

references to aggregation or group dynamics

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

phrine

-0.74

 sights

-0.71

ãĥĪ

-0.71

lihood

-0.70

©¶æ

-0.69

 Gemini

-0.65

 Antiqu

-0.63

ãĤ´ãĥ³

-0.63

ãĥīãĥ©ãĤ´ãĥ³

-0.61

CES

-0.59

POSITIVE LOGITS

regate

1.42

regation

1.18

rieved

1.10

idy

0.98

rav

0.92

art

0.85

aign

0.85

leton

0.85

arrison

0.84

reg

0.83

Activations Density 0.026%

words related to aggression

references to aggregation or group dynamics

No Comments

No Known Activations