INDEX

Explanations

instances where the word "you" is mentioned

oai_token-act-pair · gpt-3.5-turbo

references to the reader or the audience directly

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

 Immunity

-0.57

 Hours

-0.56

 srfAttach

-0.56

 None

-0.55

 Wrong

-0.55

Ĭ±

-0.54

©¶æ¥µ

-0.54

Tet

-0.54

WAYS

-0.54

Sure

-0.54

POSITIVE LOGITS

 selected

1.01

 chose

0.99

 intend

0.96

 requested

0.95

 referenced

0.92

're

0.90

 envisioned

0.87

 want

0.86

 mentioned

0.86

umbn

0.85

Activations Density 0.153%

instances where the word "you" is mentioned

references to the reader or the audience directly

No Comments

No Known Activations