INDEX

Explanations

words related to specific proper nouns, particularly names of locations and people

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to locations and names within a specific context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ffic

-0.79

Executive

-0.77

qua

-0.72

NYSE

-0.70

Canada

-0.68

India

-0.67

 Petroleum

-0.64

/-

-0.63

INT

-0.61

ICT

-0.61

POSITIVE LOGITS

heit

1.16

arest

0.97

choes

0.92

luck

0.91

bats

0.90

 Buch

0.88

ento

0.87

schild

0.87

berra

0.85

lik

0.82

Activations Density 0.011%

words related to specific proper nouns, particularly names of locations and people

proper nouns related to locations and names within a specific context

No Comments

No Known Activations