INDEX

Explanations

phrases expressing beliefs or opinions

oai_token-act-pair · gpt-3.5-turbo

assertions of belief or conviction

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

cloth

-0.71

aste

-0.71

details

-0.66

 Household

-0.64

ham

-0.62

 Navigation

-0.62

 Pavilion

-0.61

effect

-0.61

abb

-0.61

Appearance

-0.61

POSITIVE LOGITS

 passionately

0.82

ħ

0.78

§

0.75

phas

0.73

 compelled

0.70

POSE

0.70

 believe

0.69

fully

0.68

ieve

0.67

rica

0.67

Activations Density 0.044%

phrases expressing beliefs or opinions

assertions of belief or conviction

No Comments

No Known Activations