INDEX

Explanations

phrases starting with "in all" followed by different contexts or situations

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize universality or totality

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

lash

-0.70

 Schwar

-0.64

LOG

-0.64

 Fenrir

-0.63

rouse

-0.63

potion

-0.62

uced

-0.61

 Tale

-0.61

 label

-0.58

heed

-0.58

POSITIVE LOGITS

 likelihood

1.23

 seriousness

1.20

 honesty

1.20

 respects

1.19

ocating

1.15

 directions

1.09

 fairness

1.08

 phases

1.07

 contexts

1.05

 probability

1.04

Activations Density 0.050%

phrases starting with "in all" followed by different contexts or situations

phrases that emphasize universality or totality

No Comments

No Known Activations