INDEX

Explanations

phrases indicating further information or continuation in a text

oai_token-act-pair · gpt-3.5-turbo

commands or directives to read content

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

killing

-0.71

Ĭ±

-0.70

negie

-0.70

IDS

-0.70

UID

-0.65

 bandwagon

-0.64

ella

-0.62

 thwarted

-0.61

uga

-0.60

aga

-0.60

POSITIVE LOGITS

 aloud

0.93

dress

0.86

Article

0.84

 below

0.84

just

0.81

 ARTICLE

0.81

Write

0.80

 excerpts

0.80

 reviews

0.79

 about

0.79

Activations Density 0.047%

phrases indicating further information or continuation in a text

commands or directives to read content

No Comments

No Known Activations