INDEX

Explanations

terms related to different entities and their interactions, such as people, media, institutions, and coaches

oai_token-act-pair · gpt-3.5-turbo

connections and relationships between individuals and groups

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

ocious

-0.60

Lay

-0.60

Bey

-0.58

Pink

-0.57

à

-0.57

iny

-0.56

ãĢĲ

-0.56

hirt

-0.56

.<

-0.56

YC

-0.56

POSITIVE LOGITS

 surrog

0.62

 cohesion

0.57

 spectator

0.57

anwhile

0.56

 tiss

0.56

 continuum

0.55

determination

0.54

 heterogeneity

0.54

 kindred

0.54

 evolves

0.54

Activations Density 0.686%

terms related to different entities and their interactions, such as people, media, institutions, and coaches

connections and relationships between individuals and groups

No Comments

No Known Activations