INDEX

Explanations

information related to cognitive abilities and academic performance

oai_token-act-pair · gpt-3.5-turbo

numerically-based evaluations and comparisons related to cognitive and emotional assessments

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pora

-0.77

 untold

-0.72

 thous

-0.71

maid

-0.71

 Lobby

-0.70

Lago

-0.70

birds

-0.68

 billions

-0.68

 lull

-0.66

 ranc

-0.65

POSITIVE LOGITS

 questionnaire

1.35

 criteria

1.33

 tests

1.22

 exams

1.21

 checklist

1.18

 quiz

1.18

 subsc

1.15

SAT

1.14

 Examination

1.13

 criterion

1.13

Activations Density 0.355%

information related to cognitive abilities and academic performance

numerically-based evaluations and comparisons related to cognitive and emotional assessments

No Comments

No Known Activations