INDEX

Explanations

numerical values followed by units of measure

oai_token-act-pair · gpt-3.5-turbo

numerical data and statistics related to timeframes in reports

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

obi

-0.79

ãĥ¯ãĥ³

-0.79

fort

-0.75

ulla

-0.69

miss

-0.67

²¾

-0.66

vo

-0.65

oya

-0.65

»Ĵ

-0.65

sein

-0.63

POSITIVE LOGITS

 according

1.04

 reflecting

1.02

 reversing

0.99

 reaching

0.99

 dipping

0.98

 outp

0.97

 surpass

0.94

 prompting

0.94

 indicating

0.93

 suggesting

0.92

Activations Density 0.132%

numerical values followed by units of measure

numerical data and statistics related to timeframes in reports

No Comments

No Known Activations