INDEX

Explanations

phrases indicating a source of information or authority

oai_token-act-pair · gpt-3.5-turbo

phrases that start with "according to."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ograp

-0.85

oided

-0.76

ocr

-0.76

gage

-0.71

rieved

-0.70

anooga

-0.70

ennial

-0.69

eal

-0.68

 cared

-0.68

itaire

-0.65

POSITIVE LOGITS

 insiders

0.87

 sources

0.83

 experts

0.80

 STATS

0.79

reports

0.78

 Kant

0.77

 analysts

0.77

 estimates

0.74

 Fried

0.73

 reports

0.73

Activations Density 0.060%

phrases indicating a source of information or authority

phrases that start with "according to."

No Comments

No Known Activations