INDEX

Explanations

references to home runs or hitting home runs in sports contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to baseball statistics and performance

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

IFrame

Link

Negative Logits

 Imperium

-0.77

enegger

-0.77

aran

-0.71

ombat

-0.67

 Mandal

-0.66

mic

-0.66

 saddle

-0.65

MFT

-0.65

 Sahara

-0.65

ensibly

-0.64

POSITIVE LOGITS

 batted

1.09

 hitters

0.92

 hitter

0.88

 reliever

0.85

BIP

0.84

 homers

0.84

ãĥĺ

0.83

 tallied

0.79

 catcher

0.79

 batters

0.78

Activations Density 0.085%

references to home runs or hitting home runs in sports contexts

terms related to baseball statistics and performance

No Known Activations