INDEX

Explanations

references to violent actions involving firearms

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "fire" in various contexts, especially related to violent actions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Birth

-0.70

nai

-0.70

educated

-0.69

ilings

-0.64

obs

-0.63

inen

-0.62

 Brom

-0.62

elsen

-0.61

rome

-0.60

Gi

-0.60

POSITIVE LOGITS

fights

0.88

 indiscrim

0.84

 sidx

0.83

bomb

0.81

storm

0.79

geoning

0.77

[|

0.76

 exting

0.76

fight

0.74

 inflicting

0.73

Activations Density 0.026%

references to violent actions involving firearms

occurrences of the word "fire" in various contexts, especially related to violent actions

No Comments

No Known Activations