INDEX

Explanations

words that are commonly used or referenced

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate frequent or typical occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ÄŁ

-0.84

gur

-0.75

 Gareth

-0.70

 Fury

-0.66

 Majesty

-0.65

 Bagg

-0.64

udi

-0.63

onics

-0.63

stanbul

-0.62

shi

-0.62

POSITIVE LOGITS

entimes

1.00

 encountered

0.89

 known

0.86

ensical

0.85

 abbrevi

0.85

pmwiki

0.82

Used

0.82

 commonly

0.80

etheless

0.80

 referred

0.78

Activations Density 0.006%

words that are commonly used or referenced

phrases that indicate frequent or typical occurrences

No Comments

No Known Activations