INDEX

Explanations

words related to the validity, legitimacy, and legality of concepts, policies, or actions

oai_token-act-pair · gpt-3.5-turbo

terms related to legal and ethical standards of 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 @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Interstitial

-0.72

scl

-0.67

hs

-0.66

 Wonderland

-0.66

bus

-0.62

Pic

-0.62

 Sabb

-0.62

rim

-0.62

 Vent

-0.61

rive

-0.61

POSITIVE LOGITS

acies

0.91

ability

0.90

iqueness

0.88

abl

0.87

worthiness

0.86

iness

0.86

lik

0.86

 tremend

0.83

ascript

0.83

teness

0.82

Activations Density 0.053%

words related to the validity, legitimacy, and legality of concepts, policies, or actions

terms related to legal and ethical standards of validity and legitimacy

No Comments

No Known Activations