INDEX

Explanations

phrases related to specific locations or places

oai_token-act-pair · gpt-3.5-turbo

mentions of place names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Colo

-0.76

«

-0.72

Ĭ±

-0.70

¿

-0.70

=-=-=-=-

-0.69

gency

-0.66

Ľ

-0.65

ª

-0.64

 sober

-0.62

BILITIES

-0.61

POSITIVE LOGITS

mouth

0.99

pton

0.85

 Borough

0.78

wich

0.76

ople

0.75

ieri

0.75

imore

0.75

ography

0.73

aires

0.73

asonic

0.71

Activations Density 0.012%

phrases related to specific locations or places

mentions of place names

No Comments

No Known Activations