INDEX

Explanations

references to representatives or representatives in a political context

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "reps" related to repetitions in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

dar

-0.75

van

-0.75

 civilisation

-0.70

alid

-0.69

 mosaic

-0.67

 traveller

-0.66

ingen

-0.66

nes

-0.65

Bor

-0.65

 Vienna

-0.64

POSITIVE LOGITS

 reps

3.78

 Reps

2.42

 repet

1.37

RM

1.20

 spokes

1.10

 workouts

1.02

TDs

1.02

cers

1.01

 exec

0.96

 rehab

0.96

Activations Density 0.032%

references to representatives or representatives in a political context

occurrences of the term "reps" related to repetitions in various contexts

No Comments

No Known Activations