INDEX

Explanations

mentions or variations of the name "Rit" with varying activation strengths

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "rit" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

©¶æ

-0.88

ĨĴ

-0.79

«ĺ

-0.78

¶ħ

-0.78

¥ŀ

-0.75

ĻĤ

-0.64

 tender

-0.63

±

-0.62

FG

-0.61

ŃĶ

-0.60

POSITIVE LOGITS

ual

1.02

chard

0.93

chet

0.90

ique

0.88

tered

0.87

ravel

0.86

sis

0.85

ually

0.85

ters

0.84

tery

0.81

Activations Density 0.014%

mentions or variations of the name "Rit" with varying activation strengths

occurrences of the term "rit" in various contexts

No Comments

No Known Activations