INDEX

Explanations

locations or scenarios where an object or person is positioned in the middle

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "middle."

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

orthy

-0.77

ilings

-0.74

vernment

-0.70

edIn

-0.69

itsch

-0.66

/+

-0.66

cci

-0.65

ufact

-0.65

mbuds

-0.64

vec

-0.64

POSITIVE LOGITS

 layer

0.80

brow

0.71

 ages

0.71

piece

0.71

uve

0.69

 circle

0.68

pace

0.67

stad

0.66

 class

0.66

hesis

0.65

Activations Density 0.016%

locations or scenarios where an object or person is positioned in the middle

phrases that include the word "middle."

No Comments

No Known Activations