INDEX

Explanations

names of people involved in various situations or events

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ãĤ¡

-0.80

Americ

-0.67

CRE

-0.66

tml

-0.60

SAM

-0.57

 Dresden

-0.56

 Kingdoms

-0.55

soType

-0.55

category

-0.55

ãĥ¬

-0.54

POSITIVE LOGITS

 himself

0.97

 vetoed

0.95

 reportedly

0.95

 testified

0.94

 apologized

0.93

 responded

0.90

 allegedly

0.90

 replied

0.88

 denies

0.87

 countered

0.87

Activations Density 0.401%

names of people involved in various situations or events

proper nouns, particularly names of individuals

No Comments

No Known Activations