INDEX

Explanations

phrases related to losing or loss

oai_token-act-pair · gpt-3.5-turbo

repeated references to the concept of "loss."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

anwhile

-0.67

osi

-0.65

 populated

-0.63

 Austral

-0.62

 plun

-0.61

 primed

-0.58

orig

-0.58

Whe

-0.57

 BELOW

-0.57

 Municip

-0.56

POSITIVE LOGITS

itaire

0.80

 aversion

0.80

itive

0.80

luster

0.80

veyard

0.79

iem

0.75

esses

0.75

soever

0.75

 forfeit

0.70

 streaks

0.70

Activations Density 0.032%

phrases related to losing or loss

repeated references to the concept of "loss."

No Comments

No Known Activations