INDEX

Explanations

phrases related to nominations and awards

oai_token-act-pair · gpt-3.5-turbo

references to various "Best" awards and nominations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

chy

-0.76

utherford

-0.72

 enthusi

-0.70

gery

-0.69

gypt

-0.68

issance

-0.68

IGHTS

-0.68

ipt

-0.67

 inval

-0.66

tics

-0.66

POSITIVE LOGITS

seller

1.11

sell

1.08

 Best

0.95

 Worst

0.92

Best

0.91

iary

0.89

 Selling

0.88

Winner

0.86

 BaseType

0.85

 Friend

0.83

Activations Density 0.027%

phrases related to nominations and awards

references to various "Best" awards and nominations

No Comments

No Known Activations