INDEX

Explanations

mentions or references to something specific in a text

oai_token-act-pair · gpt-3.5-turbo

instances of the word "mentioned" and similar referencing phrases indicating prior discussions or points made

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ococ

-0.72

bers

-0.70

mens

-0.68

generated

-0.67

quartered

-0.62

mediated

-0.62

uga

-0.62

asm

-0.61

earances

-0.61

 rubble

-0.61

POSITIVE LOGITS

 earlier

1.72

 above

1.49

 previously

1.31

 yesterday

1.19

 before

1.17

 previous

1.10

 last

1.06

 preceding

1.03

above

0.99

before

0.98

Activations Density 0.121%

mentions or references to something specific in a text

instances of the word "mentioned" and similar referencing phrases indicating prior discussions or points made

No Comments

No Known Activations