INDEX

Explanations

hexadecimal codes for colors followed by a symbol

oai_token-act-pair · gpt-3.5-turbo

special characters and symbols 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

 contrace

-0.84

pmwiki

-0.78

milo

-0.77

 mathemat

-0.75

 recourse

-0.68

 behav

-0.67

ategory

-0.64

 compr

-0.64

isure

-0.63

 territ

-0.63

POSITIVE LOGITS

Ð°

1.07

·

1.07

Ð²

1.07

ÏĤ

1.01

Ñĭ

1.01

Ð¼

0.99

ÙĦ

0.95

×ľ

0.94

°

0.94

Ñĥ

0.92

Activations Density 0.007%

hexadecimal codes for colors followed by a symbol

special characters and symbols in the text

No Comments

No Known Activations