INDEX

Explanations

the name "ler" with varying activation levels

oai_token-act-pair · gpt-3.5-turbo

the suffix "-ler" indicating a profession or role associated with performing tasks, such as in titles or family names

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

Latin

-0.63

 behalf

-0.62

 particular

-0.62

swing

-0.61

spread

-0.59

ADRA

-0.58

 Occupations

-0.58

AQ

-0.57

awaru

-0.56

``

-0.56

POSITIVE LOGITS

gren

0.90

sonian

0.83

theless

0.82

onnaissance

0.81

uci

0.80

obyl

0.80

oyd

0.79

oser

0.78

llan

0.78

iffe

0.77

Activations Density 0.013%

the name "ler" with varying activation levels

the suffix "-ler" indicating a profession or role associated with performing tasks, such as in titles or family names

No Comments

No Known Activations