INDEX

Explanations

names of specific locations or buildings, potentially related to sports or events happening there

oai_token-act-pair · gpt-3.5-turbo

specific locations and events related to sports

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

enegger

-0.95

schild

-0.93

 Ibid

-0.81

ibliography

-0.80

ppo

-0.76

ographies

-0.76

kefeller

-0.76

 poems

-0.74

 photoc

-0.74

ographers

-0.73

POSITIVE LOGITS

 Athletic

1.14

 Wolves

1.11

 League

1.05

 Sports

1.03

 Sporting

1.01

 Esports

0.98

 league

0.98

 teammate

0.98

 Coliseum

0.98

 Lions

0.97

Activations Density 0.454%

names of specific locations or buildings, potentially related to sports or events happening there

specific locations and events related to sports

No Comments

No Known Activations