INDEX

Explanations

the Russian letter "Ð¾"

oai_token-act-pair · gpt-3.5-turbo

occurrences of a specific character or symbol in the text

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

nesday

-0.94

esville

-0.93

 Seym

-0.87

 Clover

-0.82

ometimes

-0.81

etsk

-0.77

hyde

-0.74

ologically

-0.74

raints

-0.72

secut

-0.70

POSITIVE LOGITS

ÑĢ

1.07

Ñģ

1.02

Ð½

0.96

Ñ

0.92

ÑĤ

0.88

Ðµ

0.85

Ð¾

0.81

Ì

0.80

ments

0.78

Ð

0.78

Activations Density 0.013%

the Russian letter "Ð¾"

occurrences of a specific character or symbol in the text

No Comments

No Known Activations