INDEX

Explanations

phrases related to winning or losing streaks

oai_token-act-pair · gpt-3.5-turbo

phrases related to winning or losing sequences in competitive contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

abad

-0.78

ACTED

-0.77

ãĤº

-0.73

onomic

-0.70

Lod

-0.69

gency

-0.68

resources

-0.68

poses

-0.68

hammad

-0.67

Male

-0.66

POSITIVE LOGITS

 streak

1.25

uous

0.88

 streaks

0.85

manship

0.83

icity

0.82

iness

0.80

 spanning

0.71

 breaker

0.68

ago

0.67

Activations Density 0.030%

phrases related to winning or losing streaks

phrases related to winning or losing sequences in competitive contexts

No Comments

No Known Activations