INDEX

Explanations

references to the word "giant" along with other context-specific terms

oai_token-act-pair · gpt-3.5-turbo

mentions of "giant" and its variations in contexts related to scale or size

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

say

-0.70

nikov

-0.66

nces

-0.65

ggies

-0.65

ople

-0.63

rals

-0.63

ntax

-0.62

akers

-0.62

cause

-0.61

rina

-0.61

POSITIVE LOGITS

 squid

1.16

 Squid

0.89

 bould

0.85

 leap

0.83

 Panda

0.80

ess

0.80

ape

0.78

 leaps

0.78

 strides

0.77

 Slayer

0.76

Activations Density 0.063%

references to the word "giant" along with other context-specific terms

mentions of "giant" and its variations in contexts related to scale or size

No Comments

No Known Activations