INDEX

Explanations

specific locations or points of interest

oai_token-act-pair · gpt-3.5-turbo

references to specific locations or labels associated with entities

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

href

-0.63

Army

-0.62

éĹĺ

-0.62

Loading

-0.62

Calling

-0.61

netflix

-0.59

 XCOM

-0.59

rag

-0.59

Cap

-0.58

Rail

-0.58

POSITIVE LOGITS

ORN

0.67

verages

0.66

UE

0.64

BLIC

0.64

cknowled

0.63

ourke

0.62

ñ

0.60

 guiActiveUnfocused

0.60

.,

0.60

ipolar

0.60

Activations Density 0.148%

specific locations or points of interest

references to specific locations or labels associated with entities

No Comments

No Known Activations