INDEX

Explanations

phrases related to teaching, training, or mentoring

oai_token-act-pair · gpt-3.5-turbo

elements related to judgment, evaluation, or appraisal

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 (Â£

-0.74

 maths

-0.66

 flavours

-0.64

 flavour

-0.64

 neighbourhood

-0.62

 Â£

-0.62

 Centre

-0.62

 scept

-0.62

 programmes

-0.61

 programme

-0.61

POSITIVE LOGITS

,,,,

0.78

hovah

0.69

 utilizing

0.63

 favorably

0.62

arger

0.62

 utilizes

0.61

theless

0.60

 accomplish

0.58

 thous

0.58

////////////////

0.57

Activations Density 1.551%

phrases related to teaching, training, or mentoring

elements related to judgment, evaluation, or appraisal

No Comments

No Known Activations