INDEX

Explanations

computer programming related keywords

oai_token-act-pair · gpt-3.5-turbo

elements related to titles or identifiers in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 stim

-0.94

 stun

-0.76

 compan

-0.74

 Penny

-0.72

ster

-0.70

cam

-0.69

terday

-0.68

 surve

-0.67

 livest

-0.67

panic

-0.67

POSITIVE LOGITS

II

0.93

Ģ

0.91

Legend

0.89

II

0.84

Jr

0.83

 Horus

0.82

ii

0.78

III

0.78

shapeshifter

0.77

Activations Density 0.288%

computer programming related keywords

elements related to titles or identifiers in various contexts

No Comments

No Known Activations