INDEX

Explanations

specific statements or findings within research studies

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "that."

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

andem

-0.69

gur

-0.68

 redes

-0.66

oway

-0.61

 disclaim

-0.60

Guard

-0.59

arag

-0.59

iciary

-0.57

comm

-0.57

feet

-0.56

POSITIVE LOGITS

 although

0.74

ccording

0.65

/+

0.64

 correlation

0.64

 there

0.62

 they

0.61

 contradicts

0.61

 abnormalities

0.59

bleacher

0.59

 whereas

0.59

Activations Density 0.189%

specific statements or findings within research studies

occurrences of the word "that."

No Comments

No Known Activations