INDEX

Explanations

words related to extreme behavior or intensity

oai_token-act-pair · gpt-3.5-turbo

words related to extreme behavior or mental states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Columbia

-0.69

hyd

-0.65

uter

-0.65

GH

-0.65

 Nile

-0.63

Ô

-0.61

Nag

-0.60

 Atlantis

-0.60

 SYSTEM

-0.60

pta

-0.59

POSITIVE LOGITS

iness

1.15

 craz

1.05

edly

1.00

es

1.00

iest

0.95

ese

0.92

ed

0.88

rieg

0.88

ipedia

0.87

hou

0.85

Activations Density 0.014%

words related to extreme behavior or intensity

words related to extreme behavior or mental states

No Comments

No Known Activations