INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

specific names or terms related to locations or geographic entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 respectively

-0.55

ãģ¾

-0.53

$$$$

-0.53

ãĥ¼ãĥĨ

-0.51

 âĢ¦"

-0.50

$$

-0.47

VP

-0.47

 è£ı

-0.47

SPONSORED

-0.46

,"

-0.45

POSITIVE LOGITS

 Profile

0.63

anyahu

0.57

clair

0.55

xiety

0.55

uador

0.50

omon

0.50

rimination

0.49

chwitz

0.48

utenant

0.47

ntax

0.47

Activations Density 0.583%

proper nouns

specific names or terms related to locations or geographic entities

No Comments

No Known Activations