INDEX

Explanations

terms related to enticing or alluring stimuli

oai_token-act-pair · gpt-3.5-turbo

words or terms related to entertainment and engagement activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

å£«

-0.70

 Spears

-0.67

 bearer

-0.66

 inclusive

-0.63

 scarce

-0.63

kus

-0.61

ned

-0.61

 insensitive

-0.60

 overdue

-0.60

 sterling

-0.58

POSITIVE LOGITS

ropy

1.36

ourage

1.33

rust

1.24

rance

1.23

renched

1.15

itled

1.13

osis

1.11

rench

1.10

angling

1.06

repre

1.05

Activations Density 0.021%

terms related to enticing or alluring stimuli

words or terms related to entertainment and engagement activities

No Comments

No Known Activations