INDEX

Explanations

verbs and nouns related to evaluating, judging, and assessing

oai_token-act-pair · gpt-3.5-turbo

terms related to judgment and assessment

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

dding

-0.80

zens

-0.76

vier

-0.75

seys

-0.72

LIN

-0.69

kt

-0.69

section

-0.68

ç«

-0.67

sections

-0.67

dt

-0.67

POSITIVE LOGITS

 effectiveness

1.52

worthiness

1.42

 adequ

1.32

 competence

1.30

 attractiveness

1.29

 sincerity

1.28

 merits

1.28

 efficacy

1.26

 likelihood

1.23

 viability

1.22

Activations Density 0.282%

verbs and nouns related to evaluating, judging, and assessing

terms related to judgment and assessment

No Comments

No Known Activations