INDEX

Explanations

code snippets containing functions and variable assignments

oai_token-act-pair · gpt-3.5-turbo

programming-related terms and functions within a technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

jri

-0.86

catentry

-0.77

)</

-0.77

æĪ¦

-0.74

soDeliveryDate

-0.73

BuyableInstoreAndOnline

-0.73

blogs

-0.73

AMY

-0.73

pmwiki

-0.71

è»

-0.71

POSITIVE LOGITS

->

1.21

Nil

0.90

()

0.90

=>

0.90

::

0.88

<-

0.85

0.84

 fold

0.83

IO

0.83

GHC

0.83

Activations Density 0.055%

code snippets containing functions and variable assignments

programming-related terms and functions within a technical context

No Comments

No Known Activations