INDEX

Explanations

mentions of specific mountains

oai_token-act-pair · gpt-3.5-turbo

references to specific mountains, particularly those labeled with "Mt." followed by a name

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Interstitial

-0.81

 Dragonbound

-0.78

 briefs

-0.68

submit

-0.66

EA

-0.65

crit

-0.64

Republic

-0.63

ATURE

-0.63

SHIP

-0.63

Critical

-0.63

POSITIVE LOGITS

Mt

0.92

awar

0.88

emonic

0.87

acular

0.85

unia

0.84

untary

0.79

raits

0.76

angelo

0.76

atility

0.75

aine

0.74

Activations Density 0.005%

mentions of specific mountains

references to specific mountains, particularly those labeled with "Mt." followed by a name

No Comments

No Known Activations