INDEX

Explanations

words related to receiving or getting something

oai_token-act-pair · gpt-3.5-turbo

words related to receiving or obtaining something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

iny

-0.80

pire

-0.69

hid

-0.68

pan

-0.65

apsed

-0.65

erer

-0.64

alias

-0.60

eki

-0.59

 scen

-0.58

arcity

-0.57

POSITIVE LOGITS

 preferential

1.06

 permission

1.05

 assurances

1.02

 notification

1.01

 invitations

0.96

 notice

0.95

 encouragement

0.95

 notices

0.93

 notifications

0.93

 assistance

0.93

Activations Density 0.093%

words related to receiving or getting something

words related to receiving or obtaining something

No Comments

No Known Activations