INDEX

Explanations

numbers and entities that end with the symbol "Ļ"

oai_token-act-pair · gpt-3.5-turbo

references to a specific character or symbol

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 DISTR

-0.67

Lob

-0.57

WARD

-0.56

CAST

-0.56

 blender

-0.55

 warp

-0.54

 Blend

-0.54

 disadvant

-0.54

 mathemat

-0.54

Marginal

-0.53

POSITIVE LOGITS

are

0.95

shall

0.91

were

0.90

owners

0.84

ems

0.83

ould

0.83

ttp

0.82

fore

0.79

should

0.79

aternity

0.78

Activations Density 0.120%

numbers and entities that end with the symbol "Ļ"

references to a specific character or symbol

No Comments

No Known Activations