INDEX

Explanations

mentions of specific universities, academic positions, and organizations

oai_token-act-pair · gpt-3.5-turbo

sentences that represent various academic affiliations or professional roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 sleeper

-0.74

toy

-0.68

 silhouette

-0.67

 closet

-0.67

 mummy

-0.66

 grunt

-0.65

 wardrobe

-0.65

 optional

-0.64

 phantom

-0.64

 slightest

-0.64

POSITIVE LOGITS

 Their

0.95

 Previously

0.95

 Among

0.91

 Essentially

0.91

 Though

0.90

 Specifically

0.90

 Increasing

0.89

 Those

0.89

 Conversely

0.89

 Others

0.89

Activations Density 0.206%

mentions of specific universities, academic positions, and organizations

sentences that represent various academic affiliations or professional roles

No Comments

No Known Activations