INDEX

Explanations

verbs related to success or victory

oai_token-act-pair · gpt-3.5-turbo

terms related to winning or victories in competitive scenarios

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

tan

-0.72

enegger

-0.72

asia

-0.67

ikk

-0.67

 labels

-0.67

members

-0.66

 heirs

-0.65

 successors

-0.64

ndum

-0.63

appa

-0.62

POSITIVE LOGITS

 streak

0.87

nings

0.82

ners

0.81

athan

0.81

 Coach

0.74

throp

0.74

 Defensive

0.73

 playoffs

0.73

 Lann

0.73

 Royale

0.73

Activations Density 0.045%

verbs related to success or victory

terms related to winning or victories in competitive scenarios

No Comments

No Known Activations