INDEX

Explanations

technical terms or specialized phrases related to various topics

oai_token-act-pair · gpt-3.5-turbo

terms associated with specific topics, objects, or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

racuse

-0.74

 Blaster

-0.70

\\\\\\\\

-0.65

 Courier

-0.65

 Conversation

-0.62

Adds

-0.62

 Nose

-0.60

 Temper

-0.60

 Crush

-0.59

rick

-0.59

POSITIVE LOGITS

hips

1.09

hip

0.89

 belong

0.82

hops

0.79

isphere

0.78

folk

0.77

ettings

0.74

are

0.73

mith

0.72

poons

0.71

Activations Density 0.244%

technical terms or specialized phrases related to various topics

terms associated with specific topics, objects, or entities

No Comments

No Known Activations