INDEX

Explanations

letters and symbols characterizing markup or coding languages

oai_token-act-pair · gpt-3.5-turbo

symbols or characters that signify non-standard or unconventional content formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

backer

-0.85

sburg

-0.71

iscons

-0.66

igham

-0.63

Redd

-0.62

egu

-0.61

 Maid

-0.61

CTV

-0.60

ichita

-0.59

orsi

-0.58

POSITIVE LOGITS

Ö¼

0.69

 Arrows

0.67

 Spoiler

0.62

 START

0.61

 Reson

0.61

OPT

0.59

Ð²

0.59

 arrow

0.57

apest

0.57

 Explos

0.56

Activations Density 0.164%

letters and symbols characterizing markup or coding languages

symbols or characters that signify non-standard or unconventional content formatting

No Comments

No Known Activations