INDEX

Explanations

phrases related to being caught or captured

oai_token-act-pair · gpt-3.5-turbo

phrases indicating something that has been caught on video or camera

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

idia

-0.82

enhagen

-0.75

amba

-0.72

 proxies

-0.69

arcity

-0.67

VIDIA

-0.66

amiya

-0.65

ifted

-0.65

 unification

-0.64

interstitial

-0.64

POSITIVE LOGITS

ById

0.77

tle

0.74

 Rebell

0.65

ï¸ı

0.65

EMOTE

0.64

 Sovere

0.64

Success

0.63

 Lank

0.62

 whiff

0.62

OFF

0.61

Activations Density 0.110%

phrases related to being caught or captured

phrases indicating something that has been caught on video or camera

No Comments

No Known Activations