INDEX

Explanations

instances of the word "latter" in comparison to another entity, often "former"

oai_token-act-pair · gpt-3.5-turbo

references to comparisons, particularly the terms "latter" and "former"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

bows

-0.76

jong

-0.71

rawdownloadcloneembedreportprint

-0.70

lua

-0.69

Maps

-0.67

politics

-0.67

ashes

-0.67

Magn

-0.66

 Bars

-0.65

Ped

-0.65

POSITIVE LOGITS

mentioned

0.96

 half

0.90

 stages

0.83

worldly

0.81

 part

0.81

nder

0.79

most

0.75

 halves

0.75

 category

0.75

 scenario

0.73

Activations Density 0.032%

instances of the word "latter" in comparison to another entity, often "former"

references to comparisons, particularly the terms "latter" and "former"

No Comments

No Known Activations