INDEX

Explanations

specific names or attributes related to mountains

oai_token-act-pair · gpt-3.5-turbo

mentions of "mountain."

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

BACK

-0.85

lest

-0.82

Dialogue

-0.76

encia

-0.73

NER

-0.73

âĸ¬

-0.72

uel

-0.72

tle

-0.71

orate

-0.70

âĸ¬âĸ¬

-0.70

POSITIVE LOGITS

 biking

1.13

 mountain

0.97

 Everest

0.89

 ridge

0.88

 climbers

0.88

 skiing

0.86

 footh

0.86

 Sinai

0.85

 mountains

0.82

 laure

0.81

Activations Density 0.012%

specific names or attributes related to mountains

mentions of "mountain."

No Comments

No Known Activations