INDEX

Explanations

words related to presenting arguments or opinions

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

words related to expressing opinions or points of view

oai_token-act-pair · gpt-3.5-turbo

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

verbal

-0.68

gallery

-0.63

 ILCS

-0.62

hook

-0.61

psy

-0.61

 Written

-0.58

beam

-0.57

ascript

-0.57

esc

-0.57

fell

-0.57

POSITIVE LOGITS

 against

1.22

 convinc

1.09

 passionately

1.08

 vehemently

1.04

 forcefully

0.97

 persu

0.97

against

0.94

atively

0.91

 stren

0.87

 unsuccessfully

0.84

Activations Density 0.044%

words related to presenting arguments or opinions

words related to expressing opinions or points of view

No Comments

No Known Activations