INDEX

Explanations

names of countries

oai_token-act-pair · gpt-3.5-turbo

instances of the token "la" used in various contexts

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

manship

-0.87

lers

-0.81

lessly

-0.79

ELL

-0.75

worthiness

-0.73

serious

-0.72

starter

-0.72

liners

-0.70

enegger

-0.69

leness

-0.68

POSITIVE LOGITS

uthor

1.11

uder

1.08

uren

0.92

ibrary

0.87

vel

0.85

pling

0.85

heed

0.85

igh

0.83

usk

0.81

xton

0.81

Activations Density 0.015%

names of countries

instances of the token "la" used in various contexts

No Comments

No Known Activations