INDEX

Explanations

names and locations, especially ones related to law enforcement

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals and entities, particularly focusing on the name "Sherman."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Gutenberg

-0.92

versions

-0.73

rium

-0.72

 NATO

-0.72

here

-0.72

 Yugoslav

-0.71

 Libyan

-0.70

hest

-0.67

ebook

-0.67

hire

-0.67

POSITIVE LOGITS

 Sherman

1.08

 Oaks

0.96

agements

0.88

stein

0.76

ãĥī

0.76

 Firefly

0.74

 mosqu

0.70

bach

0.70

mun

0.68

 Dome

0.68

Activations Density 0.008%

names and locations, especially ones related to law enforcement

references to specific individuals and entities, particularly focusing on the name "Sherman."

No Comments

No Known Activations