INDEX

Explanations

unrelated numbers and symbols that seem to not form a coherent pattern or theme

oai_token-act-pair · gpt-3.5-turbo

numerical data or statistics

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Courier

-0.65

lected

-0.65

 scrap

-0.60

utenant

-0.59

itives

-0.59

 Ideal

-0.59

 experien

-0.59

 antit

-0.58

utic

-0.57

 Nano

-0.57

POSITIVE LOGITS

↵

0.77

Ì¶

0.73

shapeshifter

0.69

Beta

0.69

SPONSORED

0.65

OD

0.65

ë

0.64

ãħĭ

0.64

Synopsis

0.64

ì

0.63

Activations Density 0.126%

unrelated numbers and symbols that seem to not form a coherent pattern or theme

numerical data or statistics

No Comments

No Known Activations