INDEX

Explanations

mentions of radio-related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of radio and related broadcasting terms

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

rir

-0.72

rin

-0.71

rine

-0.69

ortium

-0.68

 Staples

-0.67

rg

-0.66

 Solitaire

-0.66

 Snape

-0.65

laus

-0.65

uation

-0.65

POSITIVE LOGITS

activity

1.32

active

1.19

actively

1.16

frequency

1.16

 station

1.06

 stations

1.05

 silence

0.94

 broadcasts

0.94

 transmissions

0.93

 waves

0.90

Activations Density 0.030%

mentions of radio-related terms

mentions of radio and related broadcasting terms

No Comments

No Known Activations