INDEX

Explanations

words related to distortion or distorted forms

oai_token-act-pair · gpt-3.5-turbo

terms related to misinformation and distortion

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

ept

-0.83

Â¯Â¯Â¯Â¯

-0.81

============

-0.77

cript

-0.74

edience

-0.71

fighters

-0.69

ailable

-0.69

alions

-0.69

APTER

-0.67

Flu

-0.66

POSITIVE LOGITS

 distortions

0.96

 distorted

0.96

 distort

0.96

 distortion

0.93

 perceptions

0.75

adoes

0.74

 havoc

0.70

usional

0.69

ibly

0.68

oscope

0.64

Activations Density 0.036%

words related to distortion or distorted forms

terms related to misinformation and distortion

No Comments

No Known Activations