INDEX

Explanations

words related to military operations or command terms

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Forward."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

chens

-0.78

colo

-0.73

======

-0.71

========

-0.69

berra

-0.69

abase

-0.69

reon

-0.68

tub

-0.67

anguage

-0.66

glers

-0.66

POSITIVE LOGITS

ivity

0.74

ive

0.73

Ħ¢

0.69

 thinker

0.68

thinking

0.66

 thinkers

0.66

shore

0.66

ives

0.65

Iterator

0.64

ugal

0.63

Activations Density 0.033%

words related to military operations or command terms

occurrences of the word "Forward."

No Comments

No Known Activations