INDEX

Explanations

names or titles related to media professionals

oai_token-act-pair · gpt-3.5-turbo

terms related to commentators and pundits in media discussions

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

tering

-0.82

zman

-0.74

zik

-0.69

tered

-0.67

untu

-0.67

izont

-0.67

liness

-0.67

kamp

-0.65

esting

-0.65

enc

-0.64

POSITIVE LOGITS

 cables

0.72

 darling

0.65

ariat

0.62

 cable

0.62

onite

0.62

ial

0.61

 reckon

0.61

IAL

0.60

 analys

0.60

 feeds

0.59

Activations Density 0.121%

names or titles related to media professionals

terms related to commentators and pundits in media discussions

No Comments

No Known Activations