INDEX

Explanations

phrases related to the physical characteristics or attributes of objects

oai_token-act-pair · gpt-3.5-turbo

references to quantitative measurements or significant attributes

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

gently

-0.70

selves

-0.70

THR

-0.68

ELF

-0.65

odcast

-0.62

rahim

-0.60

shall

-0.60

din

-0.60

begin

-0.60

 Wrong

-0.59

POSITIVE LOGITS

 nature

1.06

 afforded

0.99

 similarities

0.98

 disparity

0.93

 constraints

0.92

 resemblance

0.91

 aspect

0.88

 inherent

0.88

 difference

0.88

 discrepancy

0.88

Activations Density 0.377%

phrases related to the physical characteristics or attributes of objects

references to quantitative measurements or significant attributes

No Comments

No Known Activations