INDEX

Explanations

comparisons between different values or statistics

oai_token-act-pair · gpt-3.5-turbo

phrases that involve comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ANCE

-0.75

hardt

-0.67

 Inquiry

-0.64

hack

-0.64

iband

-0.61

*/(

-0.61

Spons

-0.61

Der

-0.60

awar

-0.60

colo

-0.60

POSITIVE LOGITS

 ours

0.85

 usual

0.82

whelming

0.75

 those

0.74

 previous

0.72

average

0.70

 traditional

0.68

 peers

0.67

 typical

0.65

 mere

0.64

Activations Density 0.075%

comparisons between different values or statistics

phrases that involve comparisons

No Comments

No Known Activations