INDEX

Explanations

instances of entities related to the entertainment and pharmaceutical industries

oai_token-act-pair · gpt-3.5-turbo

occurrences of specific suffixes or word endings, particularly "er."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.88

scl

-0.76

ertodd

-0.66

 Lovecraft

-0.66

ĸļ

-0.63

ĪĴ

-0.61

âĢ¢âĢ¢

-0.60

raltar

-0.59

 disadvant

-0.58

POSITIVE LOGITS

jee

1.08

idge

0.98

unning

0.98

adish

0.97

nery

0.93

rors

0.83

lein

0.82

cial

0.82

aton

0.81

geist

0.81

Activations Density 0.096%

instances of entities related to the entertainment and pharmaceutical industries

occurrences of specific suffixes or word endings, particularly "er."

No Comments

No Known Activations