INDEX

Explanations

references to the word "Miller" in various contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Miller."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

liest

-0.85

ulates

-0.83

quo

-0.82

ulators

-0.76

rous

-0.75

ulatory

-0.75

ulating

-0.74

angular

-0.72

ulated

-0.71

ulator

-0.70

POSITIVE LOGITS

 Mayhem

0.86

ophon

0.85

 Lite

0.77

 Miller

0.77

iday

0.75

quist

0.75

hner

0.70

hound

0.69

Co

0.69

aro

0.68

Activations Density 0.039%

references to the word "Miller" in various contexts

occurrences of the name "Miller."

No Comments

No Known Activations