INDEX

Explanations

professional roles or occupations

oai_token-act-pair · gpt-3.5-turbo

references to various professional or occupational groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

tains

-0.77

Returns

-0.70

ļéĨĴ

-0.68

ľ

-0.66

 srfAttach

-0.63

ģ«

-0.62

Ħ

-0.62

potion

-0.61

ifacts

-0.61

VERTISEMENT

-0.61

POSITIVE LOGITS

 themselves

1.26

1.14

hip

1.06

']

0.94

layer

0.93

who

0.92

hips

0.91

 involved

0.91

 composing

0.87

 behind

0.86

Activations Density 0.322%

professional roles or occupations

references to various professional or occupational groups

No Comments

No Known Activations