INDEX

Explanations

phrases related to validation or legitimacy

oai_token-act-pair · gpt-3.5-turbo

words related to validity and legitimacy

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

hedon

-0.66

xual

-0.65

uration

-0.63

ynthesis

-0.61

oft

-0.61

olit

-0.60

 needless

-0.59

 unnecessary

-0.59

region

-0.58

 senseless

-0.58

POSITIVE LOGITS

 reason

0.88

ating

0.87

 excuse

0.87

 reasons

0.86

acies

0.86

ators

0.85

itimate

0.84

ator

0.81

iated

0.77

atable

0.76

Activations Density 0.104%

phrases related to validation or legitimacy

words related to validity and legitimacy

No Comments

No Known Activations