INDEX

Explanations

phrases related to sports games or events ending with the verb "loss to"

oai_token-act-pair · gpt-3.5-turbo

instances of sports-related losses and defeat

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

olor

-0.85

monitor

-0.83

pecially

-0.78

tools

-0.78

rared

-0.75

olini

-0.75

rodu

-0.71

chrome

-0.71

paren

-0.70

iott

-0.70

POSITIVE LOGITS

 finals

0.91

 Showdown

0.89

 playoff

0.89

 Finals

0.85

 playoffs

0.85

 defeat

0.84

 undefeated

0.83

 relegation

0.83

 shootout

0.82

 Champions

0.80

Activations Density 0.303%

phrases related to sports games or events ending with the verb "loss to"

instances of sports-related losses and defeat

No Comments

No Known Activations