INDEX

Explanations

phrases related to negative emotions or situations

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to objection and related actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

©¶æ

-0.78

Ô

-0.73

cale

-0.73

inventoryQuantity

-0.71

sem

-0.70

 cryst

-0.70

tnc

-0.68

 nerv

-0.64

FTWARE

-0.64

 exha

-0.63

POSITIVE LOGITS

ure

1.07

ures

1.03

urous

0.94

ions

0.92

enance

0.85

raction

0.83

ilities

0.83

ives

0.83

uation

0.78

uations

0.76

Activations Density 0.035%

phrases related to negative emotions or situations

words and phrases related to objection and related actions

No Comments

No Known Activations