INDEX

Explanations

words related to negative actions or situations

oai_token-act-pair · gpt-3.5-turbo

passive voice verbs indicating actions and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

lay

-0.69

atom

-0.67

alter

-0.67

MAT

-0.64

appers

-0.63

 Bees

-0.63

room

-0.63

thur

-0.62

aucus

-0.62

 lodge

-0.62

POSITIVE LOGITS

ĸļ

0.99

 aback

0.98

nesday

0.85

escription

0.84

ieval

0.81

 Parenthood

0.80

terness

0.79

uled

0.79

ãĤ´ãĥ³

0.77

nesses

0.77

Activations Density 0.127%

words related to negative actions or situations

passive voice verbs indicating actions and processes

No Comments

No Known Activations