INDEX

Explanations

words related to language and nationalities

oai_token-act-pair · gpt-3.5-turbo

instances of foreign characters, likely indicating different languages or coding formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ufact

-1.00

heit

-0.77

ebus

-0.71

heny

-0.69

owship

-0.67

minecraft

-0.67

bett

-0.66

bucks

-0.65

folk

-0.64

fol

-0.63

POSITIVE LOGITS

®

0.76

¾

0.74

 Archdemon

0.71

©¶æ

0.68

=-=-

0.67

å§

0.67

²

0.62

ç¥ŀ

0.61

 Torment

0.61

 Shogun

0.60

Activations Density 0.205%

words related to language and nationalities

instances of foreign characters, likely indicating different languages or coding formats

No Comments

No Known Activations