INDEX

Explanations

adjectives and phrases indicating a sense of extremity or surpassing limits

oai_token-act-pair · gpt-3.5-turbo

references to limits or constraints beyond one's control

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 differently

-0.87

 ahead

-0.73

behind

-0.72

DAY

-0.69

 alike

-0.67

soType

-0.66

 followed

-0.64

dylib

-0.63

instead

-0.61

md

-0.61

POSITIVE LOGITS

 bounds

1.28

 confines

1.13

 boundaries

1.06

 horizon

1.03

 comprehension

0.97

 threshold

0.96

 limits

0.96

 borders

0.94

 fray

0.85

 scope

0.85

Activations Density 0.158%

adjectives and phrases indicating a sense of extremity or surpassing limits

references to limits or constraints beyond one's control

No Comments

No Known Activations