INDEX

Explanations

phrases related to procedures or investigations being conducted

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "conducted" and related verbs indicating actions or activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 relevance

-0.69

pop

-0.65

ãģĦ

-0.65

 passage

-0.64

é¾įå¥ĳå£«

-0.63

inventoryQuantity

-0.62

 visibility

-0.61

 offending

-0.59

rendered

-0.59

 clipboard

-0.58

POSITIVE LOGITS

 concurrently

0.93

by

0.79

>]

0.78

 BEFORE

0.77

 anonymously

0.76

 exclusively

0.76

BY

0.75

 principally

0.75

 jointly

0.74

 electronically

0.74

Activations Density 0.206%

phrases related to procedures or investigations being conducted

occurrences of the word "conducted" and related verbs indicating actions or activities

No Comments

No Known Activations