INDEX

Explanations

words related to causing irritation or agitation

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

New Auto-Interp

Configuration

jbloom/GPT2-Small-OAI-v5-128k-attn-out-SAEs/v5_128k_layer_0

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

131,072

Data Type

torch.float32

Hook Name

blocks.0.hook_attn_out

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enhagen

-0.89

ennes

-0.88

oiler

-0.87

galitarian

-0.87

phies

-0.81

luaj

-0.80

ntil

-0.78

onse

-0.76

BIL

-0.75

rophic

-0.75

POSITIVE LOGITS

redients

0.76

fires

0.74

ãĥŁ

0.72

tenance

0.71

pmwiki

0.67

theless

0.67

argon

0.66

 Guilty

0.66

anamo

0.66

 bullets

0.66

Activations Density 0.006%

words related to causing irritation or agitation

No Comments

No Known Activations