INDEX

Explanations

phrases indicating a command or instruction being given

oai_token-act-pair · gpt-3.5-turbo

instances of the word "ordered."

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

pmwiki

-0.90

attery

-0.76

uded

-0.75

cit

-0.73

nown

-0.72

ãĤ§

-0.71

addons

-0.71

iverpool

-0.70

founded

-0.69

fact

-0.69

POSITIVE LOGITS

 execution

0.76

 executions

0.75

 inspections

0.74

 subordinates

0.73

FontSize

0.73

 corrections

0.72

 corrective

0.71

 deported

0.71

 extermin

0.71

 evacuation

0.70

Activations Density 0.071%

phrases indicating a command or instruction being given

instances of the word "ordered."

No Comments

No Known Activations