INDEX

Explanations

isolated words or parts of words ending with "ere"

oai_token-act-pair · gpt-3.5-turbo

the word "ere" in various contexts, suggesting a focus on the presence of that specific term

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

orously

-0.76

gag

-0.72

OWS

-0.71

ows

-0.70

acca

-0.68

edit

-0.67

å°Ĩ

-0.67

owed

-0.67

imates

-0.66

ished

-0.65

POSITIVE LOGITS

ndum

1.24

nda

0.95

tta

0.92

cht

0.91

ttes

0.89

lease

0.89

nder

0.86

bral

0.86

llan

0.85

cki

0.84

Activations Density 0.022%

isolated words or parts of words ending with "ere"

the word "ere" in various contexts, suggesting a focus on the presence of that specific term

No Comments

No Known Activations