INDEX

Explanations

phrases related to accusations or claims

oai_token-act-pair · gpt-3.5-turbo

the word "that" 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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

EStream

-0.82

utical

-0.79

urai

-0.79

ĸļ

-0.78

eter

-0.75

ãĤ´ãĥ³

-0.75

Laughs

-0.74

inator

-0.74

onomic

-0.74

multipl

-0.73

POSITIVE LOGITS

 they

0.82

 terrorists

0.79

 ousted

0.78

 President

0.77

 investigators

0.77

 hackers

0.76

 foreigners

0.76

 prosecutors

0.76

 vaccines

0.75

 wrongdoing

0.75

Activations Density 0.239%

phrases related to accusations or claims

the word "that" in various contexts

No Comments

No Known Activations