INDEX

Explanations

tweets or mentions on social media

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Tweet" and its variations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ccording

-0.89

EStreamFrame

-0.80

rift

-0.73

NetMessage

-0.69

undai

-0.69

cise

-0.68

å§«

-0.67

ãĥīãĥ©ãĤ´ãĥ³

-0.66

sbm

-0.65

psy

-0.64

POSITIVE LOGITS

 Tweet

1.38

Tweet

1.13

deck

1.01

edin

0.99

storms

0.98

storm

0.95

weet

0.92

 Emails

0.85

 tweet

0.82

"@

0.77

Activations Density 0.007%

tweets or mentions on social media

occurrences of the word "Tweet" and its variations

No Comments

No Known Activations