INDEX

Explanations

words related to physical damage or harm

oai_token-act-pair · gpt-3.5-turbo

references to damage 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

zsche

-0.77

zee

-0.72

rams

-0.65

KER

-0.65

zbek

-0.63

 liner

-0.61

ramid

-0.60

ILLE

-0.60

yip

-0.59

liner

-0.59

POSITIVE LOGITS

 inflicted

1.06

 damage

1.00

 mitigation

0.94

 wrought

0.93

damage

0.89

 havoc

0.82

proof

0.78

 damaged

0.78

 incurred

0.76

 limitation

0.75

Activations Density 0.026%

words related to physical damage or harm

references to damage in various contexts

No Comments

No Known Activations