INDEX

Explanations

locations or objects in a specific context

oai_token-act-pair · gpt-3.5-turbo

instances of discovery or observation in a narrative context

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

vic

-0.77

Ã±a

-0.73

nom

-0.64

 elector

-0.64

Ã±

-0.62

 endeav

-0.61

atre

-0.60

ivil

-0.60

ico

-0.59

 spirit

-0.58

POSITIVE LOGITS

ODY

0.90

EMOTE

0.80

 similarities

0.78

Suddenly

0.72

 something

0.71

IRT

0.70

 clues

0.69

 anew

0.66

 glimps

0.66

Finding

0.66

Activations Density 0.203%

locations or objects in a specific context

instances of discovery or observation in a narrative context

No Comments

No Known Activations