INDEX

Explanations

words related to targeting or aiming at something

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate intention or purpose, specifically the word "aimed."

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

wine

-0.82

amide

-0.62

 sprayed

-0.61

 Warrant

-0.61

 thereof

-0.61

check

-0.60

 Held

-0.60

kens

-0.59

plugins

-0.58

anon

-0.58

POSITIVE LOGITS

 preserving

0.91

 solving

0.89

 improving

0.87

 achieving

0.85

 maximizing

0.85

è¦ļéĨĴ

0.82

onement

0.81

 beginners

0.80

pelling

0.80

 countering

0.80

Activations Density 0.054%

words related to targeting or aiming at something

phrases that indicate intention or purpose, specifically the word "aimed."

No Comments

No Known Activations