INDEX

Explanations

adjectives with varying degrees of intensity

oai_token-act-pair · gpt-3.5-turbo

extreme descriptors and evaluative language indicating a strong opinion or condition

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 administrations

-0.57

çīĪ

-0.56

sheets

-0.56

swick

-0.53

iots

-0.53

robe

-0.52

alez

-0.52

 executes

-0.52

 throats

-0.51

 wills

-0.51

POSITIVE LOGITS

 enough

0.52

omorphic

0.52

actly

0.49

omorph

0.49

CVE

0.48

going

0.48

 dependent

0.47

 tricky

0.47

 Ichigo

0.46

isible

0.44

Activations Density 1.139%

adjectives with varying degrees of intensity

extreme descriptors and evaluative language indicating a strong opinion or condition

No Comments

No Known Activations