INDEX

Explanations

numbers or estimates with a specific value

oai_token-act-pair · gpt-3.5-turbo

terms related to estimation or evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

EED

-0.79

IGHTS

-0.75

Interstitial

-0.74

ESS

-0.73

ribution

-0.71

Va

-0.71

oret

-0.68

ous

-0.67

apers

-0.67

Cola

-0.67

POSITIVE LOGITS

peg

1.54

pole

1.13

 pegged

1.06

Peg

0.96

shit

0.90

asi

0.86

glers

0.86

holders

0.83

entary

0.82

ument

0.78

Activations Density 0.011%

numbers or estimates with a specific value

terms related to estimation or evaluation

No Comments

No Known Activations