INDEX

Explanations

It seems to focus on entities and their actions or descriptions

oai_token-act-pair · gpt-3.5-turbo

references to health and safety regulations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 advis

-0.64

 powdered

-0.63

 imitation

-0.63

jog

-0.61

itialized

-0.60

eport

-0.60

 misunder

-0.60

 puff

-0.59

 shack

-0.59

 JPEG

-0.58

POSITIVE LOGITS

 respectively

0.97

ï¸ı

0.93

_.

0.90

 Similarly

0.88

*.

0.88

ski

0.87

stellar

0.79

âĢł

0.78

thus

0.78

 Additionally

0.77

Activations Density 0.335%

It seems to focus on entities and their actions or descriptions

references to health and safety regulations

No Comments

No Known Activations