INDEX

Explanations

the prefix "im" followed by various endings or suffixes

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "im" 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 @ 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

NetMessage

-0.77

OPLE

-0.73

 Sack

-0.71

 royalties

-0.69

 slash

-0.69

 Mississ

-0.69

 Crus

-0.68

 Colleges

-0.68

 Morales

-0.67

 Waters

-0.67

POSITIVE LOGITS

unity

1.16

itates

1.10

medi

1.09

itated

1.08

balanced

1.06

imi

1.03

itating

1.03

mediate

1.00

moral

0.97

ugi

0.97

Activations Density 0.006%

the prefix "im" followed by various endings or suffixes

occurrences of the word "im" in various contexts

No Comments

No Known Activations