INDEX

Explanations

phrases or words related to threatening actions or statements made by individuals

oai_token-act-pair · gpt-3.5-turbo

instances of threats or intimidation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

gart

-0.89

æ©Ł

-0.77

rite

-0.75

mys

-0.71

_>

-0.70

served

-0.70

cedented

-0.69

prototype

-0.69

phant

-0.68

dx

-0.67

POSITIVE LOGITS

 retaliation

0.99

 repr

0.91

to

0.90

 eviction

0.90

 retribution

0.88

 violence

0.87

 expulsion

0.85

 termination

0.84

 suicide

0.82

 annihilation

0.81

Activations Density 0.039%

phrases or words related to threatening actions or statements made by individuals

instances of threats or intimidation

No Comments

No Known Activations