INDEX

Explanations

specific locations or places

oai_token-act-pair · gpt-3.5-turbo

occurrences of the verb "to be" in different forms and contexts

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sonian

-0.71

 Tsukuyomi

-0.69

prus

-0.63

inous

-0.63

 Malone

-0.62

 Hacker

-0.61

iencies

-0.60

notations

-0.58

ifa

-0.58

throp

-0.57

POSITIVE LOGITS

 headquartered

0.97

 congreg

0.90

 located

0.90

 situated

0.87

 weakest

0.85

 housed

0.79

 stationed

0.78

 resides

0.78

 nesting

0.74

 born

0.72

Activations Density 0.156%

specific locations or places

occurrences of the verb "to be" in different forms and contexts

No Comments

No Known Activations