INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

the presence of end-of-text markers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 discredited

-0.69

 unethical

-0.65

 countering

-0.64

 flawed

-0.64

isman

-0.61

 FISA

-0.61

obin

-0.61

 Turing

-0.60

 dismissing

-0.60

 rationality

-0.60

POSITIVE LOGITS

Located

0.97

 Located

0.88

 scenic

0.88

outheast

0.87

 limestone

0.85

 tourist

0.83

 sidewalks

0.81

 adjoining

0.81

 inland

0.80

lake

0.79

Activations Density 1.658%

locations or places

the presence of end-of-text markers

No Comments

No Known Activations