INDEX

Explanations

phrases indicating alignment or conformity

oai_token-act-pair · gpt-3.5-turbo

phrases indicating alignment or conformity to standards or rules

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

 livest

-0.65

 Tacoma

-0.61

iens

-0.59

odor

-0.59

 itch

-0.58

icides

-0.58

 sacrific

-0.57

 laun

-0.57

vertisements

-0.56

 strugg

-0.56

POSITIVE LOGITS

 with

0.75

arity

0.72

anthrop

0.68

 vein

0.67

With

0.67

 favour

0.66

omsky

0.63

cise

0.62

 WITH

0.62

llah

0.62

Activations Density 0.053%

phrases indicating alignment or conformity

phrases indicating alignment or conformity to standards or rules

No Comments

No Known Activations