INDEX

Explanations

mentions of the word "Sa" followed by a single character and a number

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals named "Sa" followed by additional context or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

lessly

-0.77

papers

-0.76

mercial

-0.75

tics

-0.73

breaks

-0.68

 Turing

-0.67

theless

-0.66

ancial

-0.65

ãĥ¼ãĥĨãĤ£

-0.64

å§«

-0.63

POSITIVE LOGITS

pling

1.03

adish

1.03

iva

1.01

uten

0.99

Ga

0.99

uth

0.98

igon

0.97

plings

0.96

eed

0.95

vers

0.94

Activations Density 0.011%

mentions of the word "Sa" followed by a single character and a number

references to specific individuals named "Sa" followed by additional context or titles

No Comments

No Known Activations