INDEX

Explanations

specific names or terms related to individuals, possibly in a news or informational context

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of individuals and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

OSE

-0.81

NG

-0.77

azines

-0.74

 oats

-0.74

 Tanzania

-0.71

selves

-0.71

oat

-0.71

 propri

-0.66

osity

-0.65

atches

-0.64

POSITIVE LOGITS

merga

0.85

pering

0.84

pered

0.81

SEE

0.81

formation

0.79

ptive

0.79

kell

0.78

kens

0.78

ulia

0.76

eln

0.76

Activations Density 0.027%

specific names or terms related to individuals, possibly in a news or informational context

proper nouns, particularly names of individuals and entities

No Comments

No Known Activations