INDEX

Explanations

statistics related to catches, carries, and receptions in sports

oai_token-act-pair · gpt-3.5-turbo

terms related to receptions and catches in sports

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

opian

-0.89

anian

-0.83

utral

-0.73

iazep

-0.72

oxy

-0.72

egal

-0.71

eln

-0.71

 slog

-0.70

 collabor

-0.69

 administ

-0.69

POSITIVE LOGITS

STON

0.91

AVG

0.89

 receptions

0.84

backs

0.80

 catches

0.80

yards

0.78

Ago

0.77

ãĤµ

0.77

 Receiver

0.76

 Registered

0.74

Activations Density 0.012%

statistics related to catches, carries, and receptions in sports

terms related to receptions and catches in sports

No Comments

No Known Activations