INDEX

Explanations

names containing "Prit" with varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Pritchard."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

©¶æ

-0.82

¶ħ

-0.78

«ĺ

-0.77

ĨĴ

-0.75

 sudden

-0.69

 instantaneous

-0.69

¥ŀ

-0.68

ĻĤ

-0.67

ŃĶ

-0.65

ONSORED

-0.64

POSITIVE LOGITS

ual

1.05

rit

0.96

ravel

0.95

ique

0.90

chet

0.89

ika

0.88

chard

0.86

ually

0.85

krit

0.83

iqu

0.82

Activations Density 0.007%

names containing "Prit" with varying levels of activation

occurrences of the word "Pritchard."

No Comments

No Known Activations