INDEX

Explanations

references to people with specific academic degrees, particularly Ph.D.s

oai_token-act-pair · gpt-3.5-turbo

references to individuals with Ph.D. degrees

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Bulg

-0.71

ItemTracker

-0.71

raltar

-0.69

cules

-0.67

 Isles

-0.64

éĹĺ

-0.63

 Bris

-0.63

ãĤ¡

-0.62

Io

-0.62

 Remem

-0.60

POSITIVE LOGITS

etermin

1.11

etermination

1.10

imensional

1.09

erm

1.07

ynam

1.06

ensity

1.05

ependent

1.04

iesel

1.02

iscovery

1.02

irection

1.01

Activations Density 0.047%

references to people with specific academic degrees, particularly Ph.D.s

references to individuals with Ph.D. degrees

No Comments

No Known Activations