INDEX

Explanations

references to scoring runs in sports contexts

oai_token-act-pair · gpt-3.5-turbo

mentions of scoring runs in a sports context, particularly baseball or football

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

ĨĴ

-0.87

opol

-0.86

 Males

-0.74

olia

-0.74

arist

-0.72

urers

-0.68

urnal

-0.67

arov

-0.67

ĵĺ

-0.66

ille

-0.66

POSITIVE LOGITS

swick

1.01

runner

0.81

aways

0.79

ners

0.73

gs

0.71

 batted

0.71

call

0.71

 scored

0.71

manship

0.70

runners

0.69

Activations Density 0.020%

references to scoring runs in sports contexts

mentions of scoring runs in a sports context, particularly baseball or football

No Comments

No Known Activations