INDEX

Explanations

phrases where someone is speaking in a general or non-specific manner

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a manner of discussion or analysis

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

\/\/

-0.77

ffic

-0.72

aredevil

-0.72

xtap

-0.71

Wem

-0.68

gie

-0.67

ENA

-0.67

xon

-0.66

 Survivor

-0.65

 Ballistic

-0.65

POSITIVE LOGITS

 engagements

0.83

 fluent

0.75

 impaired

0.70

 apart

0.68

 aloud

0.68

boards

0.68

 anecd

0.68

 voices

0.67

 frankly

0.67

voice

0.67

Activations Density 0.014%

phrases where someone is speaking in a general or non-specific manner

phrases that indicate a manner of discussion or analysis

No Comments

No Known Activations