INDEX

Explanations

the word "all" in different contexts

oai_token-act-pair · gpt-3.5-turbo

the term "all" in various contexts

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

IDS

-0.70

SHIP

-0.70

yip

-0.69

FH

-0.69

Few

-0.66

ciation

-0.64

rogens

-0.61

abwe

-0.61

estamp

-0.60

âĵĺ

-0.59

POSITIVE LOGITS

ocating

1.39

uding

1.09

ocated

1.03

uring

0.99

owing

0.96

edged

0.93

igators

0.90

 sorts

0.90

ergic

0.90

ocate

0.90

Activations Density 0.053%

the word "all" in different contexts

the term "all" in various contexts

No Comments

No Known Activations