INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

nouns related to locations, organizations, and products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ependent

-0.67

 enrichment

-0.63

orthy

-0.62

icter

-0.62

agnar

-0.61

inion

-0.61

 inner

-0.60

perate

-0.60

eeper

-0.60

 regression

-0.60

POSITIVE LOGITS

smanship

0.81

 sandwic

0.75

ATURES

0.74

ITS

0.69

stros

0.68

Tuc

0.67

 amassed

0.67

 Bosnia

0.67

Wars

0.67

 endured

0.64

Activations Density 0.475%

locations or places

nouns related to locations, organizations, and products

No Comments

No Known Activations