INDEX

Explanations

words related to morale, spirits, and momentum

oai_token-act-pair · gpt-3.5-turbo

terms related to morale, mood, and momentum

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

heid

-0.68

ocol

-0.68

missible

-0.66

ãĤ¨ãĥ«

-0.66

sighted

-0.66

 patents

-0.65

racist

-0.65

represent

-0.64

vertisements

-0.63

uran

-0.62

POSITIVE LOGITS

 swings

0.83

 cooker

0.77

 boost

0.75

 roller

0.74

 energ

0.74

swing

0.70

fulness

0.70

 booster

0.69

 upbeat

0.68

 overcame

0.67

Activations Density 0.175%

words related to morale, spirits, and momentum

terms related to morale, mood, and momentum

No Comments

No Known Activations