INDEX

Explanations

4-digit numbers that represent years in the 1990s

oai_token-act-pair · gpt-3.5-turbo

references to the year 1990 and its historical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

ndra

-0.76

jriwal

-0.74

 lightsaber

-0.73

hire

-0.71

llers

-0.70

holder

-0.70

Edge

-0.69

bow

-0.67

uve

-0.66

holders

-0.64

POSITIVE LOGITS

ĸļ

1.01

ãĥŁ

0.74

ãĤ¦ãĤ¹

0.73

sburg

0.67

rade

0.67

ãĥĥãĥĪ

0.65

å¹

0.65

 ï¿½ï¿½ï¿½ï¿½

0.64

aji

0.64

-'

0.63

Activations Density 0.024%

4-digit numbers that represent years in the 1990s

references to the year 1990 and its historical context

No Comments

No Known Activations