INDEX

Explanations

phrases indicating a comparison or a limit in terms of scope

oai_token-act-pair · gpt-3.5-turbo

repeated phrases beginning with "as far as."

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

Nap

-0.82

 guiActiveUn

-0.81

swick

-0.80

Nob

-0.79

TIT

-0.78

MAS

-0.78

hig

-0.76

esi

-0.76

utan

-0.76

ãĥ¼ãĥ³

-0.75

POSITIVE LOGITS

 practicable

0.95

 regards

0.93

 possible

0.91

 anyone

0.89

 anybody

0.83

 determining

0.80

 contacting

0.80

we

0.77

 respecting

0.76

 outsiders

0.75

Activations Density 0.023%

phrases indicating a comparison or a limit in terms of scope

repeated phrases beginning with "as far as."

No Comments

No Known Activations