INDEX

Explanations

phrases indicating something surpassing normal limits or expectations

oai_token-act-pair · gpt-3.5-turbo

repeated references to the concept of "beyond."

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

ICAN

-0.75

 Assass

-0.71

pmwiki

-0.71

 WATCHED

-0.68

è¦ļéĨĴ

-0.67

yah

-0.65

Ped

-0.63

footed

-0.61

cca

-0.61

 Mehran

-0.61

POSITIVE LOGITS

esis

0.83

astical

0.76

 comprehension

0.74

ordinary

0.72

 bounds

0.72

 reach

0.70

atos

0.69

 beyond

0.69

 ordinary

0.69

 Horizons

0.68

Activations Density 0.022%

phrases indicating something surpassing normal limits or expectations

repeated references to the concept of "beyond."

No Comments

No Known Activations