INDEX

Explanations

phrases indicating a reminder or a note

oai_token-act-pair · gpt-3.5-turbo

phrases that signify reminders or important notes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

Americ

-0.68

gt

-0.68

luaj

-0.66

Hon

-0.66

anooga

-0.65

Hide

-0.64

uries

-0.64

rill

-0.64

 Helpful

-0.63

Gordon

-0.63

POSITIVE LOGITS

 ONLY

0.98

 NEVER

0.92

 ALSO

0.90

 BEFORE

0.82

NOT

0.82

DID

0.81

 ALWAYS

0.81

 DOES

0.80

 MUCH

0.77

 actually

0.77

Activations Density 0.582%

phrases indicating a reminder or a note

phrases that signify reminders or important notes

No Comments

No Known Activations