INDEX

Explanations

names of prestigious awards and accolades

oai_token-act-pair · gpt-3.5-turbo

references to awards, particularly Oscars and Emmys

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

KY

-0.73

mt

-0.72

tf

-0.68

arity

-0.68

 Anon

-0.66

DF

-0.66

gm

-0.65

umbnail

-0.64

CDC

-0.63

IRT

-0.62

POSITIVE LOGITS

 nominations

1.32

 nominated

1.17

 Awards

1.17

 nominee

1.16

 nomination

1.16

 Award

1.15

 nominees

1.13

Â®

1.12

Nom

1.08

 Wilde

1.06

Activations Density 0.039%

names of prestigious awards and accolades

references to awards, particularly Oscars and Emmys

No Comments

No Known Activations