INDEX

Explanations

phrases related to credibility and trust

oai_token-act-pair · gpt-3.5-turbo

concepts related to credibility and its evaluation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Shop

-0.76

hib

-0.73

xon

-0.73

hem

-0.71

opa

-0.67

EMS

-0.67

allow

-0.66

avy

-0.66

ulton

-0.64

 Vent

-0.64

POSITIVE LOGITS

ibly

1.05

 credibility

0.92

acies

0.86

 legitimacy

0.85

 credentials

0.81

 guiActiveUn

0.75

 assurance

0.74

 guarant

0.73

itimate

0.72

 tremend

0.72

Activations Density 0.040%

phrases related to credibility and trust

concepts related to credibility and its evaluation

No Comments

No Known Activations