INDEX

Explanations

phrases related to different individuals having experience working under specific people or organizations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a context of performance evaluation or supervision

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

iterranean

-0.76

ument

-0.68

bernatorial

-0.65

ILLE

-0.62

 Edison

-0.61

-0.60

auga

-0.60

0000000000000000

-0.60

 Warsaw

-0.57

Minecraft

-0.57

POSITIVE LOGITS

pins

1.08

dogs

1.06

neath

1.04

lining

1.02

whelming

1.01

cutting

0.96

graduate

0.95

whel

0.94

lined

0.94

cuts

0.94

Activations Density 0.047%

phrases related to different individuals having experience working under specific people or organizations

phrases indicating a context of performance evaluation or supervision

No Comments

No Known Activations