INDEX

Explanations

locations like towns or cities, especially when linked to people or organizations

oai_token-act-pair · gpt-3.5-turbo

geographical locations, particularly cities and their related institutions

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

":"/

-0.75

ãĤ¢ãĥ«

-0.67

omorph

-0.64

ize

-0.64

etically

-0.63

Ĥª

-0.63

nexpected

-0.63

¼

-0.62

dule

-0.61

urgy

-0.61

POSITIVE LOGITS

 respectively

0.84

 Daughter

0.83

who

0.82

 whom

0.79

 fame

0.75

who

0.73

 plus

0.71

 whose

0.71

etc

0.69

whose

0.68

Activations Density 0.473%

locations like towns or cities, especially when linked to people or organizations

geographical locations, particularly cities and their related institutions

No Comments

No Known Activations