INDEX

Explanations

academic terms related to education, such as majors, programs, degrees, and institutions

oai_token-act-pair · gpt-3.5-turbo

academic achievements and qualifications related to education

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

 mole

-0.73

 Leviathan

-0.72

 frontman

-0.71

 smugglers

-0.64

cyn

-0.64

 appe

-0.63

 deforestation

-0.62

 eject

-0.62

 respawn

-0.62

 mete

-0.61

POSITIVE LOGITS

ploma

1.40

Students

1.18

 tuition

1.17

achelor

1.16

 Teaching

1.15

 faculty

1.15

arsity

1.13

 diploma

1.10

 vocational

1.10

Course

1.09

Activations Density 0.246%

academic terms related to education, such as majors, programs, degrees, and institutions

academic achievements and qualifications related to education

No Comments

No Known Activations