INDEX

Explanations

data visualization-related terms and multimedia content references

oai_token-act-pair · gpt-3.5-turbo

references to visual content, such as charts, videos, and articles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

entimes

-0.71

riched

-0.67

ccording

-0.62

byter

-0.61

ilver

-0.61

elsius

-0.60

aukee

-0.60

 unwelcome

-0.59

Morning

-0.59

Such

-0.58

POSITIVE LOGITS

 below

0.90

 itself

0.88

 above

0.80

liest

0.78

iest

0.76

 BEFORE

0.75

 portion

0.72

 yourself

0.71

ultimate

0.71

 behind

0.70

Activations Density 0.355%

data visualization-related terms and multimedia content references

references to visual content, such as charts, videos, and articles

No Comments

No Known Activations