INDEX

Explanations

words related to negative consequences in a social context, especially regarding morale and casualties

oai_token-act-pair · gpt-3.5-turbo

terms related to morale and casualties in military contexts

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

prints

-0.89

ridges

-0.75

odes

-0.75

ury

-0.69

sterdam

-0.65

ITNESS

-0.65

suit

-0.65

 livest

-0.65

ault

-0.64

artment

-0.64

POSITIVE LOGITS

isSpecialOrderable

0.87

 wedge

0.74

 cohesion

0.74

atform

0.74

 Haram

0.73

 underdog

0.72

 loser

0.71

izational

0.70

 morale

0.69

wcs

0.67

Activations Density 0.109%

words related to negative consequences in a social context, especially regarding morale and casualties

terms related to morale and casualties in military contexts

No Comments

No Known Activations