INDEX

Explanations

phrases related to appearance or perception

oai_token-act-pair · gpt-3.5-turbo

phrases that discuss appearance or perception

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

ItemTracker

-0.84

ifles

-0.73

PsyNetMessage

-0.71

 Bringing

-0.71

ettings

-0.71

ufact

-0.70

ordering

-0.70

 Variant

-0.69

 Introduced

-0.66

arius

-0.65

POSITIVE LOGITS

vind

0.85

 louder

0.73

 unwelcome

0.71

 stronger

0.70

 smarter

0.70

 weaker

0.67

 invincible

0.66

 uncomfortable

0.66

zynski

0.66

 disappear

0.66

Activations Density 0.065%

phrases related to appearance or perception

phrases that discuss appearance or perception

No Comments

No Known Activations