INDEX

Explanations

terms related to processes or actions

oai_token-act-pair · gpt-3.5-turbo

terms related to exclamations or emphatic expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

izont

-0.89

itism

-0.86

ACP

-0.84

bys

-0.75

Ò

-0.74

 Belg

-0.74

Ó

-0.74

kies

-0.73

DAQ

-0.68

 âĿ

-0.67

POSITIVE LOGITS

uated

0.94

uation

0.93

ulatory

0.91

uating

0.87

uate

0.83

erno

0.83

insula

0.81

uilt

0.77

ulating

0.76

uations

0.75

Activations Density 0.040%

terms related to processes or actions

terms related to exclamations or emphatic expressions

No Comments

No Known Activations