INDEX

Explanations

words related to specific place names or people's names

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names associated with a specific location or context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ONSORED

-0.84

 Scotia

-0.77

CLASSIFIED

-0.69

 Barnett

-0.68

+/-

-0.66

aneously

-0.66

Ô

-0.64

PDATE

-0.63

anwhile

-0.63

Boe

-0.62

POSITIVE LOGITS

borough

1.03

atar

1.02

erver

1.01

hart

0.98

ths

0.96

hu

0.95

haw

0.92

hin

0.91

say

0.90

ht

0.89

Activations Density 0.036%

words related to specific place names or people's names

proper nouns, particularly names associated with a specific location or context

No Comments

No Known Activations