INDEX

Explanations

technical or academic terms, possibly related to research or publications

oai_token-act-pair · gpt-3.5-turbo

terms related to academic or informational content, particularly focusing on documents, items, and products

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

aughs

-0.74

sie

-0.64

 Heck

-0.61

lihood

-0.60

 amaz

-0.59

dds

-0.58

âĦ¢:

-0.56

jah

-0.55

aug

-0.55

ghan

-0.54

POSITIVE LOGITS

 vary

0.84

 cannot

0.82

afety

0.82

are

0.82

 aren

0.81

peed

0.79

ensitive

0.76

 flowed

0.75

 tended

0.73

 must

0.73

Activations Density 0.410%

technical or academic terms, possibly related to research or publications

terms related to academic or informational content, particularly focusing on documents, items, and products

No Comments

No Known Activations