INDEX

Explanations

words related to informational content or reports, specifically in a structured format with line breaks and markers like "Ċ"

oai_token-act-pair · gpt-3.5-turbo

moments of high emotional intensity or pivotal plot revelations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 withdrawal

-0.72

 welding

-0.71

 withdraw

-0.70

illance

-0.70

 probation

-0.68

 union

-0.67

 protection

-0.67

 exchanging

-0.66

 shelling

-0.66

ewitness

-0.65

POSITIVE LOGITS

Advertisement

1.24

Favorite

1.10

Anyway

1.10

Speaking

1.01

Newsletter

1.00

SPONSORED

0.97

advertisement

0.91

Mesh

0.90

Elsewhere

0.90

Related

0.90

Activations Density 0.433%

words related to informational content or reports, specifically in a structured format with line breaks and markers like "Ċ"

moments of high emotional intensity or pivotal plot revelations

No Comments

No Known Activations