INDEX

Explanations

phrases related to using or employing something

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "utilize" and its variations

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

hood

-0.85

mad

-0.75

talk

-0.71

 compuls

-0.65

occup

-0.65

mortem

-0.64

emb

-0.64

hair

-0.62

dom

-0.61

 Guant

-0.61

POSITIVE LOGITS

 utilizes

1.06

prises

0.96

 utilization

0.95

 utilize

0.88

itures

0.81

 util

0.78

 utilized

0.78

ibles

0.78

 utilizing

0.78

edIn

0.78

Activations Density 0.009%

phrases related to using or employing something

occurrences of the word "utilize" and its variations

No Comments

No Known Activations