INDEX

Explanations

words related to a specific movie title or franchise, "Pokemon"

oai_token-act-pair · gpt-3.5-turbo

references to the character "Moke."

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

itton

-0.73

Lub

-0.70

ugal

-0.69

iltr

-0.66

ENDED

-0.64

ittal

-0.64

ittance

-0.63

inished

-0.62

arians

-0.61

orescent

-0.60

POSITIVE LOGITS

tto

0.97

ls

0.91

eper

0.89

otle

0.84

oking

0.84

rer

0.78

aways

0.78

lift

0.78

leigh

0.77

loo

0.77

Activations Density 0.011%

words related to a specific movie title or franchise, "Pokemon"

references to the character "Moke."

No Comments

No Known Activations