INDEX

Explanations

names of researchers or academics

oai_token-act-pair · gpt-3.5-turbo

proper nouns and titles of individuals, especially in academic or professional contexts

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

aepernick

-0.72

 Limbaugh

-0.69

Standing

-0.69

 Gleaming

-0.69

ramid

-0.68

HUD

-0.67

ptions

-0.67

ãĤ¦

-0.67

TRUMP

-0.67

 decency

-0.65

POSITIVE LOGITS

 doctoral

1.37

PhD

1.20

doctoral

1.16

 computational

1.15

 graduate

1.14

 undergrad

1.08

 professor

1.07

 undergraduate

1.07

 collaborators

1.06

 biomedical

1.04

Activations Density 0.132%

names of researchers or academics

proper nouns and titles of individuals, especially in academic or professional contexts

No Comments

No Known Activations