INDEX

Explanations

the word "ny" followed by a single-digit or double-digit number

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "ny"

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

EMP

-0.83

IBLE

-0.77

rador

-0.67

ibles

-0.65

Reviewed

-0.65

lessly

-0.64

iance

-0.63

ient

-0.62

OWER

-0.62

ENTION

-0.61

POSITIVE LOGITS

mph

1.00

 Mellon

0.88

acht

0.83

heter

0.82

 Giuliani

0.76

mbol

0.76

ny

0.75

eah

0.74

cling

0.73

 Barclays

0.73

Activations Density 0.022%

the word "ny" followed by a single-digit or double-digit number

occurrences of the substring "ny"

No Comments

No Known Activations