INDEX

Explanations

names or concepts that include the letters 'arg'

oai_token-act-pair · gpt-3.5-turbo

references to argumentation or the concept of argument itself

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Giov

-0.69

 fertility

-0.66

HAEL

-0.65

lihood

-0.64

Material

-0.62

 Fever

-0.62

DAY

-0.61

 tremend

-0.61

 clay

-0.61

MIT

-0.60

POSITIVE LOGITS

uably

1.28

uments

1.19

ansas

1.05

roup

1.04

emouth

0.97

arg

0.97

andom

0.90

uin

0.90

atis

0.89

rax

0.88

Activations Density 0.007%

names or concepts that include the letters 'arg'

references to argumentation or the concept of argument itself

No Comments

No Known Activations