INDEX

Explanations

phrases indicating great extent or degree

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate exceeding limits or going too far in various contexts

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

urated

-0.64

leased

-0.63

ipation

-0.60

LOT

-0.60

bed

-0.58

Nep

-0.58

urse

-0.57

leases

-0.57

ellation

-0.55

rats

-0.55

POSITIVE LOGITS

 overboard

0.89

WARD

0.79

 lengths

0.75

 towards

0.73

 toward

0.72

irtual

0.72

reaching

0.72

 beyond

0.70

 backwards

0.69

ast

0.68

Activations Density 0.056%

phrases indicating great extent or degree

phrases that indicate exceeding limits or going too far in various contexts

No Comments

No Known Activations